Developer-tools MCP server

Mnemosyne OS

Version history of discovered tools, schemas, and declared agent authority.

View watchlist

DistributionDistribution: npm
Package@mnemosyne_os/mcp
Latest analyzed version2.4.0
Current tool count31
Latest analysisNov 16, 2026
Package source

OBSERVED INTERFACE

Capability surface

Modelara’s observed declarations are shown separately from its derived effect classifications.

3 releases analyzed2.2.0 → 2.4.0Latest: 2.4.0
  • read18
  • write9
  • destructive4
  • unknown0

Unknown means Valyd could not determine the capability's effect confidently from the declared interface and static evidence. It is not treated as safe.

Package
@mnemosyne_os/mcp@2.4.0
Registry
npm
Published
Nov 14, 2026
Integrity
sha512-3QGTfazyTK7fbzxQK5yK3RK5AG3fKe2lZFNV/GYeXI1mgn4FG91iBMwx3uYPT8whZr2e7gtxtVU0YfeFxDtRNA==
Analyzer
modelara v0.1.8 · policy not recorded
Repository
https://github.com/Mnemosyne-OS/Mnemosyne-Neural-OS

31 capabilities

CapabilityTypeDerived effectFirst observed in Valyd coverageLast changed
mnemosyne_aboutmnemosyne-os-mcp:tool:mnemosyne_about
Observed interface and hashes

Re-read the server instructions: the governance tenet, the vault protection model (NORMAL and MAXIMUM, mixableWith, isolated sandbox vaults), the spine model, and the rules for an agent working on someone else's memory. Most clients show this text on connect. Call it if yours did not, or to read it again.

Input schema

{
  "properties": {},
  "type": "object"
}
sha256:3178d649f32f96ea850da50c27c062f5c20cbef86dec2f5af0587dff56ebdb67

Output schema

Not exposed in the analyzed interface.

sha256:398b676e47d03a29016ee92fe378b8b4f1b3e965390c4c64ce78d27f79df74d1

Description hash: sha256:01a0e3e4a7068a04bd9acd87193cba26d39390ac97f7a1e898a962dadd2f3ed1

Interface hash: sha256:77cb985cc9c5aba73720c8290a8927df95d85fc6370231c46effbc68d6c2909f

Other observed Modelara metadata
{
  "id": "mcp:mnemosyne-os/mnemosyne_about",
  "title": "About Mnemosyne OS",
  "source": {
    "type": "mcp",
    "server": "mnemosyne-os",
    "operation": "mnemosyne_about"
  },
  "declared": {
    "read_only": "true",
    "destructive": "unknown",
    "idempotent": "false",
    "open_world": "unknown"
  },
  "safety": {
    "idempotent": "true",
    "retry_safe": "true"
  },
  "metadata": {
    "mcp.annotations": "present"
  },
  "fingerprint": "sha256:fa31d193adefd922a5257742647c712e8152b854ac3e4f66f03d0d47decc3533"
}
toolread2.2.0No change observed
mnemosyne_agenda_addmnemosyne-os-mcp:tool:mnemosyne_agenda_add
Observed interface and hashes

Put appointments or deadlines into the user's calendar, the Agenda widget on their canvas. Use it when a conversation names a date and time to remember: a meeting, a deadline, "add this to my calendar". The app routes the write through the widget's own store, so what you file is what the user sees. Works with the app closed on a dev install (the headless daemon reads the file); an npm install has no daemon and needs the app running. To read the calendar back, change or remove an appointment, see mnemosyne_agenda_list, mnemosyne_agenda_update and mnemosyne_agenda_remove.

Input schema

{
  "properties": {
    "events": {
      "description": "One or more appointments to add.",
      "items": {
        "properties": {
          "alarm_minutes_before": {
            "description": "Minutes before \"start\" a reminder rings in the app. Omit for no reminder. 0 = at the time; 1440 = a day before.",
            "type": "number"
          },
          "all_day": {
            "description": "True for a day marker with no specific time (a deadline's date, a holiday). Default false.",
            "type": "boolean"
          },
          "description": {
            "description": "Optional notes.",
            "type": "string"
          },
          "end": {
            "description": "ISO 8601 date-time the appointment ends. Optional; omit for a point-in-time reminder or deadline rather than a block of time.",
            "type": "string"
          },
          "location": {
            "description": "Optional.",
            "type": "string"
          },
          "recurrence": {
            "description": "Omit for a one-off appointment. A recurring event keeps its day-of-month/weekday from \"start\".",
            "enum": [
              "daily",
              "weekly",
              "monthly",
              "yearly"
            ],
            "type": "string"
          },
          "start": {
            "description": "ISO 8601 date-time, e.g. \"2026-09-10T14:00:00\". No timezone offset = read as the HUMAN'S OWN machine local time, never UTC. Do not add a \"Z\" unless you mean UTC.",
            "type": "string"
          },
          "title": {
            "description": "What the appointment is, e.g. \"Dentist\" or \"Ship the report\".",
            "type": "string"
          }
        },
        "required": [
          "title",
          "start"
        ],
        "type": "object"
      },
      "type": "array"
    }
  },
  "required": [
    "events"
  ],
  "type": "object"
}
sha256:8509d1f029a60c8134129b492b110fa2f7d7888b562615508cc1e45e0c56a7d2

Output schema

Not exposed in the analyzed interface.

sha256:398b676e47d03a29016ee92fe378b8b4f1b3e965390c4c64ce78d27f79df74d1

Description hash: sha256:35fc8f30cda991a1f4c2f943a4879483e8e30d4a1cce83fc1c4c1ff0dcfadc9b

Interface hash: sha256:0847a84e3d73bd50bee088d7ec1814bcbde987a9e973a9b7c18589a0b03b53f2

Other observed Modelara metadata
{
  "id": "mcp:mnemosyne-os/mnemosyne_agenda_add",
  "title": "Add appointments",
  "source": {
    "type": "mcp",
    "server": "mnemosyne-os",
    "operation": "mnemosyne_agenda_add"
  },
  "declared": {
    "read_only": "false",
    "destructive": "false",
    "idempotent": "false",
    "open_world": "unknown"
  },
  "safety": {
    "idempotent": "false",
    "retry_safe": "false"
  },
  "metadata": {
    "mcp.annotations": "present"
  },
  "fingerprint": "sha256:35f9ce8914caefed3bf7735273e67934f10227fbfd753daf72c3f4570ca48d65"
}
toolwrite2.2.0No change observed
mnemosyne_agenda_listmnemosyne-os-mcp:tool:mnemosyne_agenda_list
Observed interface and hashes

Read the user's calendar back: the appointments in a time window, each with the id you need to change or remove it. Call this before mnemosyne_agenda_update and mnemosyne_agenda_remove, which both name appointments by id. The calendar has no archive and a removal cannot be undone. A repeating event appears once, with its cadence and the date of its next occurrence. Times in the answer are the user's machine local time. Works with the app closed on a dev install (the headless daemon reads the file); an npm install has no daemon and needs the app running. Scope agenda:read.

Input schema

{
  "properties": {
    "from": {
      "description": "ISO 8601 date-time (or epoch ms). Only appointments with an occurrence at or after this. Omitted = now.",
      "type": "string"
    },
    "include_past": {
      "default": false,
      "description": "Include appointments with no occurrence left. Default false. Those ones come back with no \"next\" date rather than with their original start dressed up as a future one.",
      "type": "boolean"
    },
    "limit": {
      "description": "Maximum appointments returned (default and ceiling: 200).",
      "type": "number"
    },
    "to": {
      "description": "ISO 8601 date-time (or epoch ms). Only appointments with an occurrence at or before this. Omitted = no upper bound, which for a repeating event means it always counts.",
      "type": "string"
    }
  },
  "type": "object"
}
sha256:691e565ad7b6809ee3c60a535b7db7f79dc65614209ec3216300d9f786fade5a

Output schema

Not exposed in the analyzed interface.

sha256:398b676e47d03a29016ee92fe378b8b4f1b3e965390c4c64ce78d27f79df74d1

Description hash: sha256:e666e73d63798d0d866dc41b7dea2d9923f6adcc543a2af2a94ee476bd6559ec

Interface hash: sha256:8d89b82e9e844e9daf79765a626d93f891279e07e20446835bbd9ddd9660e62c

Other observed Modelara metadata
{
  "id": "mcp:mnemosyne-os/mnemosyne_agenda_list",
  "title": "List appointments",
  "source": {
    "type": "mcp",
    "server": "mnemosyne-os",
    "operation": "mnemosyne_agenda_list"
  },
  "declared": {
    "read_only": "true",
    "destructive": "unknown",
    "idempotent": "false",
    "open_world": "unknown"
  },
  "safety": {
    "idempotent": "true",
    "retry_safe": "true"
  },
  "metadata": {
    "mcp.annotations": "present"
  },
  "fingerprint": "sha256:1ab26be696da05babedf820d7eb08334c667872b3d6dc8072939eeb3769c7e8b"
}
toolread2.2.0No change observed
mnemosyne_agenda_removemnemosyne-os-mcp:tool:mnemosyne_agenda_remove
Observed interface and hashes

Remove appointments from the user's calendar. Read them with mnemosyne_agenda_list first and pass the ids. This tool matches by id only, because "remove my meetings on Thursday" is how an agent removes the wrong Thursday. There is no archive: unlike a To-do task, a removed appointment is gone, so the answer names each one it removed by title and start time for the user to check. A repeating appointment is removed as a whole series, since the calendar cannot cancel a single occurrence. Works with the app closed on a dev install (the headless daemon reads the file); an npm install has no daemon and needs the app running. Scope agenda:write.

Input schema

{
  "properties": {
    "event_ids": {
      "description": "The appointment ids to remove, exactly as mnemosyne_agenda_list printed them in brackets. Up to 50.",
      "items": {
        "type": "string"
      },
      "type": "array"
    }
  },
  "required": [
    "event_ids"
  ],
  "type": "object"
}
sha256:837ec7b54c8cf8ecb837e1bb0c4ac6065fba68e17334c2c12447ab7b64fcb3df

Output schema

Not exposed in the analyzed interface.

sha256:398b676e47d03a29016ee92fe378b8b4f1b3e965390c4c64ce78d27f79df74d1

Description hash: sha256:2f17a449fc88556baf980bf6269c9fddeb9c35c5e8e46dda32e98d967641a964

Interface hash: sha256:2b50e432db6586c06b1c081b5a13e7329110ebbb80520544c8092543d08e16e7

Other observed Modelara metadata
{
  "id": "mcp:mnemosyne-os/mnemosyne_agenda_remove",
  "title": "Remove appointments",
  "source": {
    "type": "mcp",
    "server": "mnemosyne-os",
    "operation": "mnemosyne_agenda_remove"
  },
  "declared": {
    "read_only": "false",
    "destructive": "true",
    "idempotent": "false",
    "open_world": "unknown"
  },
  "safety": {
    "idempotent": "false",
    "retry_safe": "false"
  },
  "metadata": {
    "mcp.annotations": "present"
  },
  "fingerprint": "sha256:f4e2dd05973f61a9b9b191d4c08da8d0ad705ca859d90aec888a1d4294b59639"
}
tooldestructive2.2.0No change observed
mnemosyne_agenda_updatemnemosyne-os-mcp:tool:mnemosyne_agenda_update
Observed interface and hashes

Change an appointment already in the user's calendar: move it, rename it, add or drop a reminder, start or stop it repeating. It names appointments by the id from mnemosyne_agenda_list, so call that first. A field you leave out is left alone, and passing null clears it. A start or end time that cannot be read refuses the change rather than leaving the old one quietly in place. A cadence outside daily, weekly, monthly and yearly is refused rather than turned into a one-off. Works with the app closed on a dev install (the headless daemon reads the file); an npm install has no daemon and needs the app running. Scope agenda:write.

Input schema

{
  "properties": {
    "changes": {
      "description": "One entry per appointment. Up to 50 per call.",
      "items": {
        "properties": {
          "alarm_minutes_before": {
            "description": "Minutes before the start a reminder rings; null removes the reminder. Changing only this keeps the existing ring sound and volume.",
            "type": [
              "number",
              "null"
            ]
          },
          "all_day": {
            "description": "Day marker with no specific time.",
            "type": "boolean"
          },
          "description": {
            "description": "null clears it.",
            "type": [
              "string",
              "null"
            ]
          },
          "end": {
            "description": "New end, ISO 8601. null makes it a point in time rather than a block.",
            "type": [
              "string",
              "null"
            ]
          },
          "event_id": {
            "description": "The appointment id, exactly as mnemosyne_agenda_list printed it in brackets.",
            "type": "string"
          },
          "location": {
            "description": "null clears it.",
            "type": [
              "string",
              "null"
            ]
          },
          "recurrence": {
            "description": "daily / weekly / monthly / yearly, or null to STOP repeating (which also drops the series end date, since it would describe a series that no longer exists).",
            "type": [
              "string",
              "null"
            ]
          },
          "start": {
            "description": "New start, ISO 8601. No timezone offset = the HUMAN'S OWN machine local time, never UTC.",
            "type": "string"
          },
          "title": {
            "description": "New title. Cannot be emptied.",
            "type": "string"
          }
        },
        "required": [
          "event_id"
        ],
        "type": "object"
      },
      "type": "array"
    }
  },
  "required": [
    "changes"
  ],
  "type": "object"
}
sha256:52c9f830919d17ae76db9f6262fc045f6828f11d09187bc6ffef7a8dabee924f

Output schema

Not exposed in the analyzed interface.

sha256:398b676e47d03a29016ee92fe378b8b4f1b3e965390c4c64ce78d27f79df74d1

Description hash: sha256:5c14f986ff24c78cd97e3258dc8c8df13558fca483187c702a01363b40bea8d7

Interface hash: sha256:dacbac2682a2d0e8829c0641ff80a129d043a4fa333da6b1fb698dcb3c71a8c5

Other observed Modelara metadata
{
  "id": "mcp:mnemosyne-os/mnemosyne_agenda_update",
  "title": "Change an appointment",
  "source": {
    "type": "mcp",
    "server": "mnemosyne-os",
    "operation": "mnemosyne_agenda_update"
  },
  "declared": {
    "read_only": "false",
    "destructive": "true",
    "idempotent": "false",
    "open_world": "unknown"
  },
  "safety": {
    "idempotent": "false",
    "retry_safe": "false"
  },
  "metadata": {
    "mcp.annotations": "present"
  },
  "fingerprint": "sha256:2c9e1363d413d62c0fc14cc8cd36507d98b6f1a783e650537876dc8cb99d5af7"
}
tooldestructive2.2.0No change observed
mnemosyne_agent_collisionsmnemosyne-os-mcp:tool:mnemosyne_agent_collisions
Observed interface and hashes

Check whether two agent sessions are live in the same git working tree and branch right now, across every installed harness. Call it before `git add -A`, before a commit, and before a rebase: one working tree shares one git index, so a commit from one session picks up whatever the other has staged. It answers from transcript files on disk and needs neither Mnemosyne OS nor a token. Each recorded directory is resolved to its working tree first, because one `cd` into a subfolder would make two sessions in one repository look like two projects. A clean answer covers what is readable: a session whose transcripts live elsewhere does not appear at all, and sessions whose harness records no directory are listed separately as unplaceable.

Input schema

{
  "properties": {
    "project": {
      "description": "Restrict to one working tree, matched exactly (separators and case are normalised). Pass the output of `git rev-parse --show-toplevel`. Omit to check every project on the machine.",
      "type": "string"
    }
  },
  "type": "object"
}
sha256:67ec2469eac7d533a9b1ed0e9bb50bc4558d5487fffdefef0fc6db8afaf9df7e

Output schema

Not exposed in the analyzed interface.

sha256:398b676e47d03a29016ee92fe378b8b4f1b3e965390c4c64ce78d27f79df74d1

Description hash: sha256:d48feec82c88412df5d7c18bbfdc86649583d208967bf68ffc3ac59a54a1282e

Interface hash: sha256:fb8d7eda68328411506daf14330b14782abb86e176da7dd84884849eb31d2100

Other observed Modelara metadata
{
  "id": "mcp:mnemosyne-os/mnemosyne_agent_collisions",
  "title": "Check agent collisions",
  "source": {
    "type": "mcp",
    "server": "mnemosyne-os",
    "operation": "mnemosyne_agent_collisions"
  },
  "declared": {
    "read_only": "true",
    "destructive": "unknown",
    "idempotent": "false",
    "open_world": "unknown"
  },
  "safety": {
    "idempotent": "true",
    "retry_safe": "true"
  },
  "metadata": {
    "mcp.annotations": "present"
  },
  "fingerprint": "sha256:5b5dce1199ca1c74f51876b56aa0baf3124623876e23283c8cd1a4973d642b62"
}
toolread2.2.0No change observed
mnemosyne_agent_filesmnemosyne-os-mcp:tool:mnemosyne_agent_files
Observed interface and hashes

List the files other agent sessions wrote or edited recently, newest first, with the session each one came from. Paths and timestamps only. Each entry says how it is known. `recorded` means the harness logged a file-writing tool call. `from a command` means a redirection was read out of a shell command the session ran, which may never have completed. It spans every installed harness and names the agent each line came from. Use it to see what another session has already touched before you edit the same area.

Input schema

{
  "properties": {
    "contains": {
      "description": "Keep only paths containing this string. A folder name works as well as a file name.",
      "type": "string"
    },
    "limit": {
      "default": 40,
      "description": "Max files to return (default 40, max 200).",
      "type": "number"
    },
    "project": {
      "description": "Keep only files from sessions whose project path contains this string.",
      "type": "string"
    }
  },
  "type": "object"
}
sha256:742cb9b409bd76f23324aa7f2ee675995e1a4c063fe57819324656f568b058df

Output schema

Not exposed in the analyzed interface.

sha256:398b676e47d03a29016ee92fe378b8b4f1b3e965390c4c64ce78d27f79df74d1

Description hash: sha256:a17aa679817622323c899bc42ac0a4fa010e8fc37b0970fc27e7f8713c71b2f1

Interface hash: sha256:2917a4b979db2aa9287f62a40c34d2d32be754159efa83a2aa9f327e9f6888b5

Other observed Modelara metadata
{
  "id": "mcp:mnemosyne-os/mnemosyne_agent_files",
  "title": "List files an agent wrote",
  "source": {
    "type": "mcp",
    "server": "mnemosyne-os",
    "operation": "mnemosyne_agent_files"
  },
  "declared": {
    "read_only": "true",
    "destructive": "unknown",
    "idempotent": "false",
    "open_world": "unknown"
  },
  "safety": {
    "idempotent": "true",
    "retry_safe": "true"
  },
  "metadata": {
    "mcp.annotations": "present"
  },
  "fingerprint": "sha256:e2f4cd4f7f89edafee7b1a3a1e877009f0f6f67c6911dd829e87bf1e0af02197"
}
toolread2.2.0No change observed
mnemosyne_agent_listmnemosyne-os-mcp:tool:mnemosyne_agent_list
Observed interface and hashes

List the other coding-agent sessions on this machine, read from the transcript files their harnesses already write to disk. Metadata only: conversation name, project, git branch, model, last tool, how many files were touched, and when a line was last written. It reads every harness installed here, not just your own, so you can see a session from a different agent working in your repository. Use it before you touch shared state. It reports when a line was last seen and you draw the conclusion: a crashed session and an idle one fall equally silent, so nothing here can tell you an agent is working. Works with the Mnemosyne OS app closed.

Input schema

{
  "properties": {
    "limit": {
      "default": 40,
      "description": "How many transcripts to open PER HARNESS, newest first (default 40, max 200). Capping the merged total instead would let a chatty agent push a quiet one off the end, and the quiet one is the session you did not know about.",
      "type": "number"
    },
    "live_only": {
      "default": false,
      "description": "Only sessions that wrote a line recently (see the window printed in the answer). Default false, which lists the most recent sessions whether or not they moved lately.",
      "type": "boolean"
    },
    "project": {
      "description": "Keep only sessions whose project path contains this string. Pass the repository folder name to scope the answer to the repo you are working in.",
      "type": "string"
    }
  },
  "type": "object"
}
sha256:ce47040435d1ffa4cf84e33f6760d5e3154de9f3778584176239ca93447969d9

Output schema

Not exposed in the analyzed interface.

sha256:398b676e47d03a29016ee92fe378b8b4f1b3e965390c4c64ce78d27f79df74d1

Description hash: sha256:4d62df64f23d521702c412d1338a03c93147973ce8e2a8e7debcc6d0bfcbb354

Interface hash: sha256:d7793c10680e0f8e73a20675ec72929f78d74a234aa534b3e700b104a4876a2b

Other observed Modelara metadata
{
  "id": "mcp:mnemosyne-os/mnemosyne_agent_list",
  "title": "List coding agents",
  "source": {
    "type": "mcp",
    "server": "mnemosyne-os",
    "operation": "mnemosyne_agent_list"
  },
  "declared": {
    "read_only": "true",
    "destructive": "unknown",
    "idempotent": "false",
    "open_world": "unknown"
  },
  "safety": {
    "idempotent": "true",
    "retry_safe": "true"
  },
  "metadata": {
    "mcp.annotations": "present"
  },
  "fingerprint": "sha256:0f96f6e6bec8b0f398670c0fb3ceca0a0d78081040a929411482b60859c28274"
}
toolread2.2.0No change observed
mnemosyne_cockpit_updatemnemosyne-os-mcp:tool:mnemosyne_cockpit_update
Observed interface and hashes

Update your own status card on the user's canvas, the cockpit. Call it when you start a task ("working", with a short title and status), when you need the user ("waiting"), when you are stuck ("blocked"), and when you finish ("done"). "waiting" and "blocked" make the card pulse and the taskbar flash. Send "waiting" IN THE SAME TURN as the question you ask the user, right before you stop: the harness only knows to say "waiting" for a permission prompt, and a question asked in the chat is otherwise just the end of a turn — the user never sees that you are waiting for them. Your "waiting" survives the harness saying the turn ended. You declare the state; the app prints it next to the time since your last call, so keep calling at real milestones or the card goes quiet. The answer carries any message the user left on your card. Read it and act on it. Needs the app window open. This is a card, so nothing is stored in memory.

Input schema

{
  "properties": {
    "attention": {
      "default": false,
      "description": "Ask for the human's eye even in \"working\"/\"done\" (the card pulses, the taskbar flashes). \"waiting\" and \"blocked\" ask on their own.",
      "type": "boolean"
    },
    "desktop": {
      "description": "The desktop this session should put its cards on, by name, exactly as the human wrote it. Send it once when they tell you; the card stays there for the rest of the conversation without you repeating it. The match is exact apart from case, and an unknown or duplicated name is reported back with the names that exist rather than guessed at. Leave it out and the card goes to whichever desktop answers for this project, which is what most sessions want.",
      "type": "string"
    },
    "detail": {
      "description": "Up to 4 short lines under the status (files, a branch, a count). Optional.",
      "items": {
        "type": "string"
      },
      "type": "array"
    },
    "session": {
      "description": "Only when the harness publishes no CLAUDE_CODE_SESSION_ID: a stable id for this conversation, reused on every call.",
      "type": "string"
    },
    "state": {
      "description": "\"working\" = on it; \"waiting\" = you asked the human something and stopped; \"done\" = the task is finished; \"blocked\" = you cannot continue without them; \"closed\" = this conversation is over (removes the card).",
      "enum": [
        "working",
        "waiting",
        "done",
        "blocked",
        "closed"
      ],
      "type": "string"
    },
    "status": {
      "description": "One line: what you are doing right now, or what you need. 160 characters max.",
      "type": "string"
    },
    "title": {
      "description": "What this conversation is about, in a few words (the card's title). Give it at least once; later calls may omit it.",
      "type": "string"
    }
  },
  "required": [
    "state"
  ],
  "type": "object"
}
sha256:5961cee2a6026483736f487477d98b60871da385c1341ad59c93d232d70e9e42

Output schema

Not exposed in the analyzed interface.

sha256:398b676e47d03a29016ee92fe378b8b4f1b3e965390c4c64ce78d27f79df74d1

Description hash: sha256:2ddac09bbc9efd49220a27242db510596cb2af48293f9d3da7189ea5e24416e4

Interface hash: sha256:bd59c9548e5014d9c6034f0e1ef9582b9da7782ef26b3bf42344c15275320239

Other observed Modelara metadata
{
  "id": "mcp:mnemosyne-os/mnemosyne_cockpit_update",
  "title": "Update your status card",
  "source": {
    "type": "mcp",
    "server": "mnemosyne-os",
    "operation": "mnemosyne_cockpit_update"
  },
  "declared": {
    "read_only": "false",
    "destructive": "false",
    "idempotent": "false",
    "open_world": "unknown"
  },
  "safety": {
    "idempotent": "false",
    "retry_safe": "false"
  },
  "metadata": {
    "mcp.annotations": "present"
  },
  "fingerprint": "sha256:67f8c56b1acb5eb7f12b76f919165dbc71de8cba3d2773be2ac9ca739c9f546b"
}
toolwrite2.2.0No change observed
mnemosyne_dream_bridgesmnemosyne-os-mcp:tool:mnemosyne_dream_bridges
Observed interface and hashes

List the connections the Dream State engine found between memories while the machine was idle. Each bridge links two memories, sometimes from different vaults, with a Dream Bridge Score, a raw cosine similarity, and a short excerpt of both sides. A side marked "inferred" was matched on its id and spine type in the only vault that fits (older scans did not record the vault): likely, not certain. Use it to surface associations the memory made on its own, to audit whether they are insightful or noise, or to seed creative exploration. An empty list is normal and means the engine has not produced bridges yet. It runs on idle time, when the user enables it in Settings.

Input schema

{
  "properties": {
    "chronicle_id": {
      "description": "Only bridges touching this chronicle id (either endpoint).",
      "type": "number"
    },
    "include_unresolved": {
      "description": "Also list bridges where a memory could not be found (ambiguous or missing side). Default false: they are left out and counted.",
      "type": "boolean"
    },
    "limit": {
      "default": 50,
      "description": "Max bridges to return, strongest (highest dbs) first (default: 50, server cap: 200).",
      "type": "number"
    },
    "min_dbs": {
      "description": "Only bridges with a Dream Bridge Score at or above this value (0-1).",
      "type": "number"
    },
    "session_id": {
      "description": "Restrict to one dream scan session.",
      "type": "string"
    }
  },
  "type": "object"
}
sha256:cacb820e5dbe9b1d90df6ba2bda41a93c052bc2a81ec0bcb683823ce55471f63

Output schema

Not exposed in the analyzed interface.

sha256:398b676e47d03a29016ee92fe378b8b4f1b3e965390c4c64ce78d27f79df74d1

Description hash: sha256:509fe6c00f404016f37fc10da739b1a768c14bd454edfbfc423f20e903391cdd

Interface hash: sha256:f4e42c335fe56813a3123510fff8518b0fcf14fef50a51da7330f674670d41a6

Other observed Modelara metadata
{
  "id": "mcp:mnemosyne-os/mnemosyne_dream_bridges",
  "title": "List dream bridges",
  "source": {
    "type": "mcp",
    "server": "mnemosyne-os",
    "operation": "mnemosyne_dream_bridges"
  },
  "declared": {
    "read_only": "true",
    "destructive": "unknown",
    "idempotent": "false",
    "open_world": "unknown"
  },
  "safety": {
    "idempotent": "true",
    "retry_safe": "true"
  },
  "metadata": {
    "mcp.annotations": "present"
  },
  "fingerprint": "sha256:503bcc2ab1f0ec8a817d1136b5d10754cd803cb317f0601300c615487bc97011"
}
toolread2.2.02.3.0
mnemosyne_git_logmnemosyne-os-mcp:tool:mnemosyne_git_log
Observed interface and hashes

Read recent commits from the git repository this Mnemosyne OS is configured to read. Read-only. Each commit carries an 8-character hash, the subject line, the author and the date, newest first. The path is set on the app side, so this cannot be pointed at another checkout, and it needs the monorepo:read scope. When either is missing the answer names what is missing, rather than returning an empty list that would read as "no commits". Use mnemosyne_memory_query for the reasoning behind a change.

Input schema

{
  "properties": {
    "limit": {
      "default": 20,
      "description": "Number of commits to return (default: 20)",
      "type": "number"
    },
    "since": {
      "default": "30 days ago",
      "description": "Time range (e.g. \"7 days ago\", \"2024-01-01\")",
      "type": "string"
    }
  },
  "type": "object"
}
sha256:c6897737bbb82548cae9c30485e24722c1022b6877a0bbcede4cde86a1b74b0a

Output schema

Not exposed in the analyzed interface.

sha256:398b676e47d03a29016ee92fe378b8b4f1b3e965390c4c64ce78d27f79df74d1

Description hash: sha256:e7493c68e8eefb170483b2f86bc2035b1f8465a23bd07242bb9e1de2fec514f5

Interface hash: sha256:c6333109131abe0ddba23f8a6841a6bfc89d0d887f472a7f5a3cb0b63a1e4990

Other observed Modelara metadata
{
  "id": "mcp:mnemosyne-os/mnemosyne_git_log",
  "title": "Read git history",
  "source": {
    "type": "mcp",
    "server": "mnemosyne-os",
    "operation": "mnemosyne_git_log"
  },
  "declared": {
    "read_only": "true",
    "destructive": "unknown",
    "idempotent": "false",
    "open_world": "unknown"
  },
  "safety": {
    "idempotent": "true",
    "retry_safe": "true"
  },
  "metadata": {
    "mcp.annotations": "present"
  },
  "fingerprint": "sha256:995218d07d607e52fef6d4d2423aa7175c2bc1468e2f53add534be12e38934e8"
}
toolread2.2.0No change observed
mnemosyne_mac_update_checkmnemosyne-os-mcp:tool:mnemosyne_mac_update_check
Observed interface and hashes

macOS only. Check whether a newer Mnemosyne OS is published and what an update would do, without writing anything. It reads the installed version in /Applications, the latest release manifest (latest-mac.yml), the macOS version, whether the app is running, and whether the key files exist (never their content). Works with the app closed. Call it before mnemosyne_mac_update_install and show its answer to the person.

Input schema

{
  "properties": {},
  "type": "object"
}
sha256:3178d649f32f96ea850da50c27c062f5c20cbef86dec2f5af0587dff56ebdb67

Output schema

Not exposed in the analyzed interface.

sha256:398b676e47d03a29016ee92fe378b8b4f1b3e965390c4c64ce78d27f79df74d1

Description hash: sha256:398cb7a8d0bcf96f4da6d4bbcafc010aca9261a59e768ecfd1ce0da9d825610d

Interface hash: sha256:22e6c2d35fac29542827d2fd86cffe949e8a3d96a48ea14a1a8472e4dfabc5ab

Other observed Modelara metadata
{
  "id": "mcp:mnemosyne-os/mnemosyne_mac_update_check",
  "title": "Check for a Mac update",
  "source": {
    "type": "mcp",
    "server": "mnemosyne-os",
    "operation": "mnemosyne_mac_update_check"
  },
  "declared": {
    "read_only": "true",
    "destructive": "unknown",
    "idempotent": "false",
    "open_world": "unknown"
  },
  "safety": {
    "idempotent": "true",
    "retry_safe": "true"
  },
  "metadata": {
    "mcp.annotations": "present"
  },
  "fingerprint": "sha256:6cf4fdccd5d8eca39bbdaa205815ade914ae9c43214eef2e72f6a0248ed7a989"
}
toolread2.4.02.4.0
mnemosyne_mac_update_installmnemosyne-os-mcp:tool:mnemosyne_mac_update_install
Observed interface and hashes

macOS only. Install the latest Mnemosyne OS release in place of the one in /Applications, the way a careful person would. It downloads the dmg and checks its size and sha512 against latest-mac.yml, copies keys.enc, keys-transfer.json and model-config.json to ~/.mnemosyne/update-backups/<time>/, asks the app to quit and waits for its own shutdown (it never kills it), copies the new app beside the old one, verifies its signature, swaps them (the old app goes to the backup folder) and relaunches. It stops at the first refusal and says which. Works with the app closed. After it, macOS asks the PERSON for Keychain access: tell them to click Always Allow; never click it for them. Needs the exact version mnemosyne_mac_update_check reported and confirm: true.

Input schema

{
  "properties": {
    "confirm": {
      "description": "true, once the person agreed to the update.",
      "type": "boolean"
    },
    "version": {
      "description": "The version mnemosyne_mac_update_check reported as the latest release, exactly.",
      "type": "string"
    }
  },
  "required": [
    "version",
    "confirm"
  ],
  "type": "object"
}
sha256:056c9ea9b647997f7da55f36500e30b07514bec49588e60f0a59544cda4cfe42

Output schema

Not exposed in the analyzed interface.

sha256:398b676e47d03a29016ee92fe378b8b4f1b3e965390c4c64ce78d27f79df74d1

Description hash: sha256:3e175ba24caff96c99c5589d17cbe28f3dd311ebd6bcdc4fa0b8b3fbc96b3963

Interface hash: sha256:a7e92cc212936c91b77f89e3727a54b46681354d644f8a971c26a7954e2a913d

Other observed Modelara metadata
{
  "id": "mcp:mnemosyne-os/mnemosyne_mac_update_install",
  "title": "Install the Mac update",
  "source": {
    "type": "mcp",
    "server": "mnemosyne-os",
    "operation": "mnemosyne_mac_update_install"
  },
  "declared": {
    "read_only": "false",
    "destructive": "true",
    "idempotent": "false",
    "open_world": "unknown"
  },
  "safety": {
    "idempotent": "false",
    "retry_safe": "false"
  },
  "metadata": {
    "mcp.annotations": "present"
  },
  "fingerprint": "sha256:2f348c691e913baf6fa3b4252583a3d12988902093fff9444d54765e34863d09"
}
tooldestructive2.4.02.4.0
mnemosyne_memory_askmnemosyne-os-mcp:tool:mnemosyne_memory_ask
Observed interface and hashes

Ask a question and get a prose answer grounded in the vault, plus the memories it drew on. It runs the full local RAG pipeline: deeper retrieval, lexical fusion and a re-rank. That also makes it the better retriever, so reach for it whenever you need to find something, and read the Sources list even if you ignore the prose. It is at its best on why, who and how questions that span many memories. It takes up to about 30 seconds. The prose is a model rewording of the sources, so quote the sources instead, or fetch them with mnemosyne_memory_query. Check them before you trust the answer.

Input schema

{
  "properties": {
    "question": {
      "description": "A question in plain language, the way you would ask a colleague who knows the material. Be specific.",
      "type": "string"
    },
    "vault": {
      "default": "DEV",
      "description": "Vault to reason over, case-insensitive. Default for this deployment: \"DEV\". A vault this server is not scoped for is refused.",
      "type": "string"
    }
  },
  "required": [
    "question"
  ],
  "type": "object"
}
sha256:cacd45599fae73e58bc949f9bc962b7075c5d8b08bd3e618d7fb5632bc91c50e

Output schema

Not exposed in the analyzed interface.

sha256:398b676e47d03a29016ee92fe378b8b4f1b3e965390c4c64ce78d27f79df74d1

Description hash: sha256:8930851a8c7cd59982a48996528795a54e40043a79ed526e3e2af21c2def2d4f

Interface hash: sha256:ad69f2b079a8d8fed6e45ba4512bc2d5448ffc136ee8196a67719bfbb61f3743

Other observed Modelara metadata
{
  "id": "mcp:mnemosyne-os/mnemosyne_memory_ask",
  "title": "Ask the vault",
  "source": {
    "type": "mcp",
    "server": "mnemosyne-os",
    "operation": "mnemosyne_memory_ask"
  },
  "declared": {
    "read_only": "true",
    "destructive": "unknown",
    "idempotent": "false",
    "open_world": "unknown"
  },
  "safety": {
    "idempotent": "true",
    "retry_safe": "true"
  },
  "metadata": {
    "mcp.annotations": "present"
  },
  "fingerprint": "sha256:cdf998f284a3d4bb48c215cfd3a006cbb441593387321fa4c0b8f546fc220c0e"
}
toolread2.2.0No change observed
mnemosyne_memory_ingestmnemosyne-os-mcp:tool:mnemosyne_memory_ingest
Observed interface and hashes

Write a memory into a vault: a decision, an architecture note, a debug finding, a session summary. It is stored permanently and indexed for every future agent to retrieve. Use it at the end of a meaningful work session, or whenever a decision is reached that someone will want to recall later.

Input schema

{
  "properties": {
    "content": {
      "description": "The content to store. Markdown works. Write it self-contained, and say WHY the decision was made rather than only what was decided.",
      "type": "string"
    },
    "spine_type": {
      "default": "NOTE",
      "description": "Semantic type of the content. ARCHITECTURE is boosted heavily in code-scope queries: use it for design docs, big-picture decisions and structural choices. DECISION for narrower trade-offs. BUGFIX and DEBUG for incident learnings. SESSION for where you left off. FEATURE for new capabilities. NOTE for the rest.",
      "enum": [
        "DECISION",
        "ARCHITECTURE",
        "DEBUG",
        "BUGFIX",
        "FEATURE",
        "NOTE",
        "SESSION",
        "RESONANCE",
        "CUSTOM"
      ],
      "type": "string"
    },
    "vault": {
      "default": "DEV",
      "description": "Target vault TOKEN: the folder name uppercased, with spaces and hyphens as underscores. Default for this deployment: \"DEV\". Scoped tokens, which are configuration rather than a census: DEV, PERSONAL, SOCIAL. Ingest is permanent, so confirm the vault exists with mnemosyne_vault_list before writing anywhere new.",
      "type": "string"
    }
  },
  "required": [
    "content"
  ],
  "type": "object"
}
sha256:7a513e7229578bedb45721bc2d246678f3462d06f12baa394d98df9c178d808c

Output schema

Not exposed in the analyzed interface.

sha256:398b676e47d03a29016ee92fe378b8b4f1b3e965390c4c64ce78d27f79df74d1

Description hash: sha256:8fd371cc299b2a24b6400c2979718535831b3dee5155c4192b236f7062f627bd

Interface hash: sha256:986fbe11c0d155e277103618270621319c54d417b3289a5001c104ad40d4fc1c

Other observed Modelara metadata
{
  "id": "mcp:mnemosyne-os/mnemosyne_memory_ingest",
  "title": "Save a memory",
  "source": {
    "type": "mcp",
    "server": "mnemosyne-os",
    "operation": "mnemosyne_memory_ingest"
  },
  "declared": {
    "read_only": "false",
    "destructive": "false",
    "idempotent": "false",
    "open_world": "unknown"
  },
  "safety": {
    "idempotent": "false",
    "retry_safe": "false"
  },
  "metadata": {
    "mcp.annotations": "present"
  },
  "fingerprint": "sha256:200836effdee6ad26113b260fc2ac5d264ce5f79ae08c8728651802caa38dde5"
}
toolwrite2.2.02.4.0
mnemosyne_memory_querymnemosyne-os-mcp:tool:mnemosyne_memory_query
Observed interface and hashes

Search a vault and get the matching memories back word for word: notes, code, decisions, session summaries, commit history. Nothing is rewritten, so use this when you need the source text itself, to quote it or to write documentation from it. Ranking is vector similarity fused with a local BM25 channel, weighted by spine type. To FIND something rather than quote it, prefer mnemosyne_memory_ask, even when you only want its sources. It retrieves deeper and re-ranks, so it catches rare literal terms this tool misses: proper nouns, identifiers, product names. The score is not a confidence value. A hit and a miss come back with similar numbers, so judge the text, not the number beside it.

Input schema

{
  "properties": {
    "limit": {
      "default": 10,
      "description": "Max number of results (default: 10, max: 50)",
      "type": "number"
    },
    "max_content_chars": {
      "default": 600,
      "description": "Per-memory snippet size in characters (default 600). Each result is cut to this length, with a hint about its full size. Raise it past 2000 when you need whole files, and watch your context window.",
      "type": "number"
    },
    "query": {
      "description": "The search query. Be specific. Examples: \"why we moved off Postgres\", \"the retry logic in the payment worker\", \"notes from the March planning session\".",
      "type": "string"
    },
    "spine_type_filter": {
      "description": "Optional whitelist of spine types. Use [\"ARCHITECTURE\"] for design docs, [\"GIT\"] for commit history, [\"BUGFIX\",\"DEBUG\"] for incident knowledge, [\"SOURCE_CODE\"] for code only. Omit it to get every type, with the SOURCE_CODE scope weighting deciding the ranking.",
      "items": {
        "type": "string"
      },
      "type": "array"
    },
    "vault": {
      "default": "DEV",
      "description": "Vault TOKEN to search, case-insensitive. A token is the folder name uppercased, with spaces and hyphens as underscores. This deployment's default is \"DEV\". It is also scoped for: PERSONAL, SOCIAL. That list is configuration rather than a census, so call mnemosyne_vault_list to see which of them exist on this machine. Anything outside the list is refused.",
      "type": "string"
    }
  },
  "required": [
    "query"
  ],
  "type": "object"
}
sha256:bbd2f7858e3f129ac815f3b9b962754fca320ddff4389eaba90cf30cff3dbd4b

Output schema

Not exposed in the analyzed interface.

sha256:398b676e47d03a29016ee92fe378b8b4f1b3e965390c4c64ce78d27f79df74d1

Description hash: sha256:defa79f95fa882c91c132c9a25818b4e7d367fe30ff44b76e551b61be2664320

Interface hash: sha256:3fd4398ba5d59e8b524c95903148ec1fb78d158138fe7fc30716f72c6a4bb0d5

Other observed Modelara metadata
{
  "id": "mcp:mnemosyne-os/mnemosyne_memory_query",
  "title": "Search chronicles",
  "source": {
    "type": "mcp",
    "server": "mnemosyne-os",
    "operation": "mnemosyne_memory_query"
  },
  "declared": {
    "read_only": "true",
    "destructive": "unknown",
    "idempotent": "false",
    "open_world": "unknown"
  },
  "safety": {
    "idempotent": "true",
    "retry_safe": "true"
  },
  "metadata": {
    "mcp.annotations": "present"
  },
  "fingerprint": "sha256:a380ef91a31772a0f81775f1838347c1cc8428923ef8082a4e3556c3dff5541c"
}
toolread2.2.02.4.0
mnemosyne_pheme_radarmnemosyne-os-mcp:tool:mnemosyne_pheme_radar
Observed interface and hashes

Read what the user's Pheme radar last found: fresh threads in the subreddits and Hacker News queries they watch. Each thread carries a topic score, and a tier once they have run the Mnemosyne pass. The tier says how much substance they can bring to that thread. The answer leads with when the scan ran, because it is only as fresh as the last time they opened Pheme and pressed Scan. Use it to find threads to draft a reply for; the user posts it. "No radar" means no scan has been projected yet, so ask them to open Pheme and scan. Needs the app running, though Pheme itself can be closed.

Input schema

{
  "properties": {
    "limit": {
      "default": 30,
      "description": "How many threads at most (default 30, max 100).",
      "type": "number"
    },
    "tier": {
      "description": "Only threads of this tier. Omit for all (unranked ones included, last).",
      "enum": [
        "high",
        "mid",
        "low"
      ],
      "type": "string"
    }
  },
  "type": "object"
}
sha256:cd351a8e2e46b20e615c2f036c83d5d5634a160f027657a51e84357a0b47e409

Output schema

Not exposed in the analyzed interface.

sha256:398b676e47d03a29016ee92fe378b8b4f1b3e965390c4c64ce78d27f79df74d1

Description hash: sha256:e0b078959de569ca9a464e919cec057ec6d5eae266c81c45a6c98750adee0249

Interface hash: sha256:5cab67fc629bc1ee0d41eae4b91254743872ee8a442b49d3318816e15e059be8

Other observed Modelara metadata
{
  "id": "mcp:mnemosyne-os/mnemosyne_pheme_radar",
  "title": "Read the Pheme radar",
  "source": {
    "type": "mcp",
    "server": "mnemosyne-os",
    "operation": "mnemosyne_pheme_radar"
  },
  "declared": {
    "read_only": "true",
    "destructive": "unknown",
    "idempotent": "false",
    "open_world": "unknown"
  },
  "safety": {
    "idempotent": "true",
    "retry_safe": "true"
  },
  "metadata": {
    "mcp.annotations": "present"
  },
  "fingerprint": "sha256:ca56a57bd5f4ac7c9f637329ce09aaecb72c413525f6bc70040ed8e13fffcad6"
}
toolread2.2.0No change observed
mnemosyne_pheme_watchmnemosyne-os-mcp:tool:mnemosyne_pheme_watch
Observed interface and hashes

Add a subreddit, a Hacker News search query or a topic to the user's Pheme radar, or take one off. Pheme is their reputation cartridge: it finds fresh threads worth a genuine reply. The lists belong to them. An operation that would empty one is refused. Every operation reports its own outcome: done, already there, not there, refused. The user gets a receipt in Pheme naming this agent and what changed. Use it when a conversation finds a community worth watching. It scans nothing and posts nothing; the user posts. Needs the app running, though Pheme itself can be closed.

Input schema

{
  "properties": {
    "ops": {
      "description": "The changes, in order. Up to 50.",
      "items": {
        "properties": {
          "kind": {
            "description": "\"sub\" = a subreddit (with or without r/); \"topic\" = an expertise topic that scores every network's radar; \"hnQuery\" = a Hacker News search query.",
            "enum": [
              "sub",
              "topic",
              "hnQuery"
            ],
            "type": "string"
          },
          "op": {
            "enum": [
              "add",
              "remove"
            ],
            "type": "string"
          },
          "value": {
            "description": "The subreddit name, topic or query. 120 characters max.",
            "type": "string"
          }
        },
        "required": [
          "op",
          "kind",
          "value"
        ],
        "type": "object"
      },
      "type": "array"
    }
  },
  "required": [
    "ops"
  ],
  "type": "object"
}
sha256:91a4e561cff93307503857a34a10eb12ac4548199387ac21d0e429db8e3d1048

Output schema

Not exposed in the analyzed interface.

sha256:398b676e47d03a29016ee92fe378b8b4f1b3e965390c4c64ce78d27f79df74d1

Description hash: sha256:258a663e7c54b1e9bcbfd6cae096bbeb4a1d246c5a2edc8020cda5497a477487

Interface hash: sha256:2c35b0ac1ad04c4c6cddbde8c861c69191abfcc3f76a9cc31889f2dbe3d8602b

Other observed Modelara metadata
{
  "id": "mcp:mnemosyne-os/mnemosyne_pheme_watch",
  "title": "Edit the Pheme watchlist",
  "source": {
    "type": "mcp",
    "server": "mnemosyne-os",
    "operation": "mnemosyne_pheme_watch"
  },
  "declared": {
    "read_only": "false",
    "destructive": "false",
    "idempotent": "false",
    "open_world": "unknown"
  },
  "safety": {
    "idempotent": "false",
    "retry_safe": "false"
  },
  "metadata": {
    "mcp.annotations": "present"
  },
  "fingerprint": "sha256:b3302b6c84c31b7563936b212fa836fe676f440be4c380c947cca971e1853e44"
}
toolwrite2.2.0No change observed
mnemosyne_position_getmnemosyne-os-mcp:tool:mnemosyne_position_get
Observed interface and hashes

Read where one resonance was left: the phase, and the free-text note an agent or the cockpit wrote when it stopped. Read-only. It returns the resonance id, when it was saved, the spine type of the memory, and the whole note. An id nothing was ever saved under answers in plain words and points at mnemosyne_position_update, so a blank history and a failed call do not look alike. Call mnemosyne_resonance_list when you do not know the id.

Input schema

{
  "properties": {
    "resonance_id": {
      "description": "ID of the resonance, for example \"checkout-rewrite\" or \"v2-migration\".",
      "type": "string"
    }
  },
  "required": [
    "resonance_id"
  ],
  "type": "object"
}
sha256:305b54f82aceffae59e38aaf12412661817863c7d53c1e5855c309debd989dfb

Output schema

Not exposed in the analyzed interface.

sha256:398b676e47d03a29016ee92fe378b8b4f1b3e965390c4c64ce78d27f79df74d1

Description hash: sha256:90fbb15af6ffb5e9925c09af7fbb8ab6c72d6996b517727207c0a3bbbdb1f7ef

Interface hash: sha256:b8e9842d37abf37895cbb1fa55efa3b25fd493ffdc3dfddd41ca6fc91c809783

Other observed Modelara metadata
{
  "id": "mcp:mnemosyne-os/mnemosyne_position_get",
  "title": "Read a resonance position",
  "source": {
    "type": "mcp",
    "server": "mnemosyne-os",
    "operation": "mnemosyne_position_get"
  },
  "declared": {
    "read_only": "true",
    "destructive": "unknown",
    "idempotent": "false",
    "open_world": "unknown"
  },
  "safety": {
    "idempotent": "true",
    "retry_safe": "true"
  },
  "metadata": {
    "mcp.annotations": "present"
  },
  "fingerprint": "sha256:0f8b7de5c5237e8c8eed7bd5e5165251b35d2120646125ef5d2f837f3cdfe7dc"
}
toolread2.2.0No change observed
mnemosyne_position_updatemnemosyne-os-mcp:tool:mnemosyne_position_update
Observed interface and hashes

Record where you left off on a resonance: phase, current state, next steps. Call it at the end of a session. It is stored as a DECISION memory in the vault.

Input schema

{
  "properties": {
    "phase": {
      "description": "Current phase label (e.g. \"Phase 52\", \"v1.1.0 release\")",
      "type": "string"
    },
    "position": {
      "description": "Description of the current position / what was done / what is next",
      "type": "string"
    },
    "resonance_id": {
      "description": "ID of the resonance to update",
      "type": "string"
    }
  },
  "required": [
    "resonance_id",
    "position",
    "phase"
  ],
  "type": "object"
}
sha256:e5b3b64be127e7bb7e9c2683e5025c346a3502e55eb88cb76b5683aff8c235b7

Output schema

Not exposed in the analyzed interface.

sha256:398b676e47d03a29016ee92fe378b8b4f1b3e965390c4c64ce78d27f79df74d1

Description hash: sha256:455306a63596c42a489d8a4117e02245a04058bff4737801f24afd820b1b331a

Interface hash: sha256:215c4f27f73c4386d610643de7b8c34863caf0cf69ab4f8c4514302ff43771be

Other observed Modelara metadata
{
  "id": "mcp:mnemosyne-os/mnemosyne_position_update",
  "title": "Record where you left off",
  "source": {
    "type": "mcp",
    "server": "mnemosyne-os",
    "operation": "mnemosyne_position_update"
  },
  "declared": {
    "read_only": "false",
    "destructive": "false",
    "idempotent": "false",
    "open_world": "unknown"
  },
  "safety": {
    "idempotent": "false",
    "retry_safe": "false"
  },
  "metadata": {
    "mcp.annotations": "present"
  },
  "fingerprint": "sha256:3d0c32b862cda2cc2c455e02aac22f8264ccdc25495a59530d3a8b58fa575c66"
}
toolwrite2.2.0No change observed
mnemosyne_resonance_listmnemosyne-os-mcp:tool:mnemosyne_resonance_list
Observed interface and hashes

List the resonances recorded in the default vault. A resonance is a workspace that tracks one ongoing project. Each entry carries its id, its last phase, how many minutes ago it moved, and the id of the memory behind it. Read-only. There is no status filter, so you get every resonance the scan matched in one vault, from at most 30 candidates. An empty result is written in words and means no resonance has been recorded yet. Read one in full with mnemosyne_position_get, or write a new one with mnemosyne_position_update.

Input schema

{
  "properties": {},
  "type": "object"
}
sha256:3178d649f32f96ea850da50c27c062f5c20cbef86dec2f5af0587dff56ebdb67

Output schema

Not exposed in the analyzed interface.

sha256:398b676e47d03a29016ee92fe378b8b4f1b3e965390c4c64ce78d27f79df74d1

Description hash: sha256:cc25d466d90785835274d3e32b726bf4bc0642d370a8fa6e8817b527ac016caa

Interface hash: sha256:169fdf1ff9ee397dea5ce1386ddac18cee2e58b40db0e34b07bb83614e768a69

Other observed Modelara metadata
{
  "id": "mcp:mnemosyne-os/mnemosyne_resonance_list",
  "title": "List resonances",
  "source": {
    "type": "mcp",
    "server": "mnemosyne-os",
    "operation": "mnemosyne_resonance_list"
  },
  "declared": {
    "read_only": "true",
    "destructive": "unknown",
    "idempotent": "false",
    "open_world": "unknown"
  },
  "safety": {
    "idempotent": "true",
    "retry_safe": "true"
  },
  "metadata": {
    "mcp.annotations": "present"
  },
  "fingerprint": "sha256:dda4e2ee8a2be3b405f464135bdfed459c1961cbaf85cbc62b4d42a87bdfa21b"
}
toolread2.2.0No change observed
mnemosyne_review_getmnemosyne-os-mcp:tool:mnemosyne_review_get
Observed interface and hashes

Read one review as it stands: its state (waiting / approved / changes / rejected), every answer so far, and its publishing state. Does not wait. Needs the Mnemosyne OS app RUNNING: on macOS its window may be closed (the app stays up and the host reads the file); on Windows and Linux closing the last window quits the app, so it must be open. Scope review:ask.

Input schema

{
  "properties": {
    "id": {
      "description": "The review id.",
      "type": "string"
    }
  },
  "required": [
    "id"
  ],
  "type": "object"
}
sha256:825759c03e63ba246e169f46670d4875a20a0105d2912a4992a8d8c3dd50c1db

Output schema

Not exposed in the analyzed interface.

sha256:398b676e47d03a29016ee92fe378b8b4f1b3e965390c4c64ce78d27f79df74d1

Description hash: sha256:3537a39ae37c0d331e8c481f9e8eb4c6bf7bdd1e16bb4a9520f8745487cebc64

Interface hash: sha256:e034e2dfc4f0be0bbb00fb15d57a534612dd93d33be542d3910f720bfe29b344

Other observed Modelara metadata
{
  "id": "mcp:mnemosyne-os/mnemosyne_review_get",
  "title": "Read a review",
  "source": {
    "type": "mcp",
    "server": "mnemosyne-os",
    "operation": "mnemosyne_review_get"
  },
  "declared": {
    "read_only": "true",
    "destructive": "unknown",
    "idempotent": "false",
    "open_world": "unknown"
  },
  "safety": {
    "idempotent": "true",
    "retry_safe": "true"
  },
  "metadata": {
    "mcp.annotations": "present"
  },
  "fingerprint": "sha256:811bfdd74672ada7ed44db6cc2f1fda48cc074ab7a572281a7f93bda91799cef"
}
toolread2.3.02.3.0
mnemosyne_review_openmnemosyne-os-mcp:tool:mnemosyne_review_open
Observed interface and hashes

Ask the human to review a video or image you produced. It shows in the review panel of Mnemosyne OS and, when the human paired the review bot, on their phone (Telegram) with Approve / Change / Reject buttons. Returns the review id: pass it to mnemosyne_review_wait. The file must be an absolute path inside ~/.mnemosyne/generations, ~/.mnemosyne/voiceovers or ~/.mnemosyne/outbox (copy a render there first); anything else is refused with OUTSIDE_SEND_FOLDERS. For a retake of a review, pass its id as parentId: the panel shows the versions side by side. Needs the Mnemosyne OS app RUNNING: on macOS its window may be closed (the app stays up and the host reads the file); on Windows and Linux closing the last window quits the app, so it must be open. Scope review:ask.

Input schema

{
  "properties": {
    "fps": {
      "description": "Optional: the video frame rate (e.g. 30, from the render settings). The panel steps frame by frame at this rate and the notes of the human come back with exact frame numbers. Without it the panel assumes 30 and the human can change it.",
      "type": "number"
    },
    "parentId": {
      "description": "Optional: the review this one is a retake of.",
      "type": "string"
    },
    "path": {
      "description": "Absolute path of the video or image (.mp4 .mov .webm .png .jpg .webp .gif).",
      "type": "string"
    },
    "project": {
      "description": "Optional: the film project id this render belongs to (the Films window of Mnemosyne OS, e.g. \"stop-aux-gadgets\"; `avatar-film status <film>` prints it). The review then shows on the board of that film. A retake (parentId) keeps the project of its parent.",
      "type": "string"
    },
    "question": {
      "description": "Optional: what you want them to look at (max 600 characters).",
      "type": "string"
    },
    "title": {
      "description": "Short title the human sees, e.g. \"Alberto birthday v2\" (max 120 characters).",
      "type": "string"
    }
  },
  "required": [
    "path",
    "title"
  ],
  "type": "object"
}
sha256:948dad377e4ad45c59b22eb41a1b4b03f25a3db16545cdae42e38ed45338e0c9

Output schema

Not exposed in the analyzed interface.

sha256:398b676e47d03a29016ee92fe378b8b4f1b3e965390c4c64ce78d27f79df74d1

Description hash: sha256:e9fe8f3529114979871d42cecff24f6f6f613a3e05d37a9d91ae94c37c27da05

Interface hash: sha256:6840d1f92782da8daa224e5a147ba7224b3f8de2be42bc111862f4f7860caeac

Other observed Modelara metadata
{
  "id": "mcp:mnemosyne-os/mnemosyne_review_open",
  "title": "Ask the human to review a video",
  "source": {
    "type": "mcp",
    "server": "mnemosyne-os",
    "operation": "mnemosyne_review_open"
  },
  "declared": {
    "read_only": "false",
    "destructive": "false",
    "idempotent": "false",
    "open_world": "unknown"
  },
  "safety": {
    "idempotent": "false",
    "retry_safe": "false"
  },
  "metadata": {
    "mcp.annotations": "present"
  },
  "fingerprint": "sha256:e160f0318b06c8c395e9c9ecf783cc88da65109978701eec901d91a09f746a13"
}
toolwrite2.3.02.3.0
mnemosyne_review_waitmnemosyne-os-mcp:tool:mnemosyne_review_wait
Observed interface and hashes

Wait for the human's answers to ONE review (from mnemosyne_review_open), from the app or from their phone. Returns every answer given after `after`, oldest first: verdict approve / changes / reject, or null for a plain comment (the human can send frame notes alone, without a verdict), plus the text (a voice note arrives transcribed) and the second of the video it points at when they gave one. From the app an answer can carry frame notes: exact frame numbers (counted at the fps shown), each with what to fix there. Pass `after` (the createdAt of the review, or the `after` the last result told you) so an answer given before you started waiting is not missed. ONE CALL WAITS AT MOST ~45 s, so it never outlives your client's own tool timeout. When it returns STILL WAITING, nothing was decided: call it again with the same id and the same after, in a loop, for as long as you mean to listen. The human sees in the panel whether an agent is listening. Needs the Mnemosyne OS app RUNNING: on macOS its window may be closed (the app stays up and the host reads the file); on Windows and Linux closing the last window quits the app, so it must be open. Scope review:ask.

Input schema

{
  "properties": {
    "after": {
      "description": "Epoch ms: only answers strictly after this count. Default: now.",
      "type": "number"
    },
    "id": {
      "description": "The review id.",
      "type": "string"
    },
    "timeout_seconds": {
      "description": "How long this one call waits, default and max 45 (raised only by MNEMO_REVIEW_WAIT_MAX_S for a client known to wait longer). To listen longer, call again on STILL WAITING.",
      "type": "number"
    }
  },
  "required": [
    "id"
  ],
  "type": "object"
}
sha256:e1a2f9de4c3e2eb57ae10ddeb151d2bcd7c5dd808d454e5232915f6da0fb8060

Output schema

Not exposed in the analyzed interface.

sha256:398b676e47d03a29016ee92fe378b8b4f1b3e965390c4c64ce78d27f79df74d1

Description hash: sha256:afa18724894f57a238461b24cb23990f831facffa1411b1b0229b4abd474ca27

Interface hash: sha256:4f039b99a296eaae664e15cf534798b36e99c54bbd92c0e92dbf60ddd53f65b4

Other observed Modelara metadata
{
  "id": "mcp:mnemosyne-os/mnemosyne_review_wait",
  "title": "Wait for the answer to a review",
  "source": {
    "type": "mcp",
    "server": "mnemosyne-os",
    "operation": "mnemosyne_review_wait"
  },
  "declared": {
    "read_only": "false",
    "destructive": "false",
    "idempotent": "false",
    "open_world": "unknown"
  },
  "safety": {
    "idempotent": "false",
    "retry_safe": "false"
  },
  "metadata": {
    "mcp.annotations": "present"
  },
  "fingerprint": "sha256:fda7ffab923254fa2df2cb1bcc08ac3800ad9c76551435999d13ad9d897a773b"
}
toolwrite2.3.02.3.0
mnemosyne_spine_assignmentsmnemosyne-os-mcp:tool:mnemosyne_spine_assignments
Observed interface and hashes

See how Mnemosyne classified its memories. It returns the memory-to-spine assignments for one vault, newest first, and the per-spine counts for the whole vault. Use it to check whether memories landed in the right spines, to read a vault's composition at a glance, or to find the taxon ids to pass as spine_type_filter in mnemosyne_memory_query. Set include_taxonomy to also get the global spine tree.

Input schema

{
  "properties": {
    "include_taxonomy": {
      "default": false,
      "description": "Also return the global spine taxonomy tree (natures → sub-spines).",
      "type": "boolean"
    },
    "limit": {
      "default": 25,
      "description": "Max assignments returned (default: 25 to stay context-friendly, server cap: 500).",
      "type": "number"
    },
    "offset": {
      "default": 0,
      "description": "Pagination offset (default: 0).",
      "type": "number"
    },
    "spine_type": {
      "description": "Restrict the assignment page to one spine taxon id (e.g. \"DOCUMENT\", \"GIT\"). Counts stay whole-vault.",
      "type": "string"
    },
    "vault": {
      "default": "DEV",
      "description": "Vault to inspect, case-insensitive. Default for this deployment: \"DEV\".",
      "type": "string"
    }
  },
  "type": "object"
}
sha256:2aca328576fe6e0ee03dd319d70f60ab1610c5e2b6798b3321bc0e1213d5a94b

Output schema

Not exposed in the analyzed interface.

sha256:398b676e47d03a29016ee92fe378b8b4f1b3e965390c4c64ce78d27f79df74d1

Description hash: sha256:294ed83730fcbf8128e61388ec7d0039a419cc7c0cf88363f49b28ff842993f9

Interface hash: sha256:492963cb08042d062463fd05a433ceb1a99bfa30c96744ad4391eece2053759e

Other observed Modelara metadata
{
  "id": "mcp:mnemosyne-os/mnemosyne_spine_assignments",
  "title": "List spine assignments",
  "source": {
    "type": "mcp",
    "server": "mnemosyne-os",
    "operation": "mnemosyne_spine_assignments"
  },
  "declared": {
    "read_only": "true",
    "destructive": "unknown",
    "idempotent": "false",
    "open_world": "unknown"
  },
  "safety": {
    "idempotent": "true",
    "retry_safe": "true"
  },
  "metadata": {
    "mcp.annotations": "present"
  },
  "fingerprint": "sha256:8db40169cb22e8fb89eaf154926afd98142b539c165fa55108d0bc01ca6e54bf"
}
toolread2.2.0No change observed
mnemosyne_telegram_sendmnemosyne-os-mcp:tool:mnemosyne_telegram_send
Observed interface and hashes

Send one local file (video, photo, audio, pdf) to the user's Telegram home chat — the same Hermes bot channel they already configured. Use it when a render or export finished on disk and the user is not at the machine: pass the absolute path, optionally a short caption. The recipient is NEVER chosen by you: only TELEGRAM_HOME_CHANNEL from the assistant .env. Caps: 48 MB, allow-listed types (mp4/mov/webm/mkv, jpg/png/webp/gif, wav/mp3/ogg, pdf). Needs the Mnemosyne OS app RUNNING: on macOS its window may be closed (the app stays up and the host reads the file); on Windows and Linux closing the last window quits the app, so it must be open. Refusals: NO_TELEGRAM (bot or home chat missing), TOO_LARGE, BAD_TYPE, NOT_A_FILE, BAD_REQUEST.

Input schema

{
  "properties": {
    "caption": {
      "description": "Optional caption shown under the media (max 1024 characters).",
      "type": "string"
    },
    "path": {
      "description": "Absolute path of the file on this machine (e.g. a finished MP4). Relative paths are refused.",
      "type": "string"
    }
  },
  "required": [
    "path"
  ],
  "type": "object"
}
sha256:54fdb3b51f68fd4a2d13627e9d0427601e1fec372ebf14a79ba144185993a037

Output schema

Not exposed in the analyzed interface.

sha256:398b676e47d03a29016ee92fe378b8b4f1b3e965390c4c64ce78d27f79df74d1

Description hash: sha256:29c13062638708c3521a5bfb4ff7d39f5d760c426c3f1d2c3534b55069c2cd7d

Interface hash: sha256:bca303eee0e4350d11b65eaf37e4f4fd606b31dc994682113d139bae43fe4cd5

Other observed Modelara metadata
{
  "id": "mcp:mnemosyne-os/mnemosyne_telegram_send",
  "title": "Send a file to Telegram home",
  "source": {
    "type": "mcp",
    "server": "mnemosyne-os",
    "operation": "mnemosyne_telegram_send"
  },
  "declared": {
    "read_only": "false",
    "destructive": "false",
    "idempotent": "false",
    "open_world": "unknown"
  },
  "safety": {
    "idempotent": "false",
    "retry_safe": "false"
  },
  "metadata": {
    "mcp.annotations": "present"
  },
  "fingerprint": "sha256:8aafcf5323a5d3e22246ff247f2b502576d045a64c936d17c6f5f72f0b274590"
}
toolwrite2.3.02.3.0
mnemosyne_todo_addmnemosyne-os-mcp:tool:mnemosyne_todo_add
Observed interface and hashes

Put tasks into the user's To-do backlog, the To-do widget on their canvas, in order and optionally under named steps. Use it when a conversation has settled what to do. Name the destination list in "list": call once without it to get the existing lists back on a LIST_NOT_FOUND answer, or pass create_list: true to make a new one. Always name a list, and file one plan into ONE list, its steps as "group". Write each text as a short title a person can say aloud (about 80 characters); the detail goes in "description". A rule, a trap or a decision is not something to do: file it with kind "note", so it stays as context and counts in no total. On a recent app your tasks wait in « To triage » until the human keeps them, and each gets a number (#42) shown by mnemosyne_todo_list. The app routes the write through the widget's own store, so what you file is what the user sees. Works with the app closed on a dev install (the headless daemon reads the file); an npm install has no daemon and needs the app running. To read the backlog back or change what is in it, see mnemosyne_todo_list, mnemosyne_todo_update and mnemosyne_todo_categories.

Input schema

{
  "properties": {
    "color": {
      "description": "Hex colour (#rrggbb) for a list being created. Optional; the host picks the next swatch otherwise.",
      "type": "string"
    },
    "create_list": {
      "default": false,
      "description": "Create the list named in \"list\" when it does not exist. Default false, so an agent does not invent lists in someone's backlog without saying so.",
      "type": "boolean"
    },
    "list": {
      "description": "Displayed name of the destination list, case-insensitive: \"Today\", \"WIP\", or a list the user created. Omitted, it goes to the first original list. An unknown name with create_list false is refused, and the answer gives the names that exist.",
      "type": "string"
    },
    "tasks": {
      "description": "The tasks, in execution order. Each item is a string, or an object: \"text\" (a short title), \"group\" (the STEP it belongs to, e.g. \"Step 1 · Mockup\"; tasks with the same group are shown under one header), \"description\" (the detail, 4000 max), \"kind\" (\"note\" for context that is not a thing to do), \"status\" (\"doing\", \"waiting\" or \"someday\"), \"priority\" (\"high\" or \"low\"), \"attachments\" (absolute paths of files or folders the task is about, or {\"path\", \"kind\": \"file\"|\"dir\", \"label\"}). 300 characters max per text; a longer text is filed shortened and the answer says so.",
      "items": {
        "anyOf": [
          {
            "type": "string"
          },
          {
            "properties": {
              "attachments": {
                "items": {
                  "anyOf": [
                    {
                      "type": "string"
                    },
                    {
                      "properties": {
                        "kind": {
                          "enum": [
                            "file",
                            "dir"
                          ],
                          "type": "string"
                        },
                        "label": {
                          "type": "string"
                        },
                        "path": {
                          "type": "string"
                        }
                      },
                      "required": [
                        "path"
                      ],
                      "type": "object"
                    }
                  ]
                },
                "type": "array"
              },
              "description": {
                "type": "string"
              },
              "group": {
                "type": "string"
              },
              "kind": {
                "enum": [
                  "note"
                ],
                "type": "string"
              },
              "priority": {
                "enum": [
                  "high",
                  "low"
                ],
                "type": "string"
              },
              "status": {
                "enum": [
                  "doing",
                  "waiting",
                  "someday"
                ],
                "type": "string"
              },
              "text": {
                "type": "string"
              }
            },
            "required": [
              "text"
            ],
            "type": "object"
          }
        ]
      },
      "type": "array"
    }
  },
  "required": [
    "tasks"
  ],
  "type": "object"
}
sha256:4bb0fa2f081aaafc9d702034147c75847bb85a94ef2f0a114f250094c4a844ab

Output schema

Not exposed in the analyzed interface.

sha256:398b676e47d03a29016ee92fe378b8b4f1b3e965390c4c64ce78d27f79df74d1

Description hash: sha256:1ed95da77982f954908144963eebf7256b8cab440c269007238be95f536c818e

Interface hash: sha256:6af115d3cc05d949474088e8a03b47e7c4a99e9b7733f4aa198b1aa6e823c017

Other observed Modelara metadata
{
  "id": "mcp:mnemosyne-os/mnemosyne_todo_add",
  "title": "Add To-do tasks",
  "source": {
    "type": "mcp",
    "server": "mnemosyne-os",
    "operation": "mnemosyne_todo_add"
  },
  "declared": {
    "read_only": "false",
    "destructive": "false",
    "idempotent": "false",
    "open_world": "unknown"
  },
  "safety": {
    "idempotent": "false",
    "retry_safe": "false"
  },
  "metadata": {
    "mcp.annotations": "present"
  },
  "fingerprint": "sha256:f12ce3973896246481ee92c54e63a2a35d04e8c612a4b1dd0a724f1dfd99c981"
}
toolwrite2.2.0No change observed
mnemosyne_todo_categoriesmnemosyne-os-mcp:tool:mnemosyne_todo_categories
Observed interface and hashes

Manage the lists of the user's To-do backlog: create one, rename or recolour one, remove an empty one. It is separate from mnemosyne_todo_update because these change the shape of a workspace rather than the work in it. Two refusals are worth knowing before you call. A list that still holds tasks is never removed, and you are told how many are in the way: move them first, since the app will not pick a destination on someone's behalf. The three original lists can be renamed, never removed, because their contents are what make the file readable. Works with the app closed on a dev install (the headless daemon reads the file); an npm install has no daemon and needs the app running. Scope todo:write.

Input schema

{
  "properties": {
    "ops": {
      "description": "The changes, applied in order - so you can create a list and rename another in one call.",
      "items": {
        "properties": {
          "color": {
            "description": "Hex colour (#rrggbb). Optional; the host picks the next swatch for a new list.",
            "type": "string"
          },
          "list": {
            "description": "list.edit / list.remove: which list, by displayed name or key.",
            "type": "string"
          },
          "name": {
            "description": "list.create: the name. list.edit: the new name; passing an empty string on one of the three ORIGINAL lists gives it back its localised default name.",
            "type": "string"
          },
          "op": {
            "description": "list.create = a new empty list. list.edit = rename and/or recolour. list.remove = delete an EMPTY, non-original list.",
            "enum": [
              "list.create",
              "list.edit",
              "list.remove"
            ],
            "type": "string"
          }
        },
        "required": [
          "op"
        ],
        "type": "object"
      },
      "type": "array"
    }
  },
  "required": [
    "ops"
  ],
  "type": "object"
}
sha256:889e0787cdb1130f06d4443f2ee2ac8f9f8e1a2eba97d3be5a4dc09dcd30f031

Output schema

Not exposed in the analyzed interface.

sha256:398b676e47d03a29016ee92fe378b8b4f1b3e965390c4c64ce78d27f79df74d1

Description hash: sha256:df8114037e0bc88323dcba699fb57fe25144d441ec7df2bd7b7af0a109b35530

Interface hash: sha256:5ece90b2583fcd2c95d5a692bd6df3778533f13a24f3462227ba06a9579410de

Other observed Modelara metadata
{
  "id": "mcp:mnemosyne-os/mnemosyne_todo_categories",
  "title": "List To-do lists",
  "source": {
    "type": "mcp",
    "server": "mnemosyne-os",
    "operation": "mnemosyne_todo_categories"
  },
  "declared": {
    "read_only": "true",
    "destructive": "unknown",
    "idempotent": "false",
    "open_world": "unknown"
  },
  "safety": {
    "idempotent": "true",
    "retry_safe": "true"
  },
  "metadata": {
    "mcp.annotations": "present"
  },
  "fingerprint": "sha256:ab71216d51a5afc00b7bfab4dea43c0653e6669667b4d43b17a10c0605193d07"
}
toolread2.2.0No change observed
mnemosyne_todo_listmnemosyne-os-mcp:tool:mnemosyne_todo_list
Observed interface and hashes

Read the user's To-do backlog back: the lists that exist (the person may call a list a category or a project) and the tasks in them, each with the id (and the short number, #42) you need to change it, its status, priority, whether it is a note or waits in triage, and the names of its attached files. Call this before mnemosyne_todo_update, which names tasks by id or number. A phrase like "delete the task about the invoice" reads perfectly and can still match the wrong task. This is also the way to answer "what is on my plate", or to check whether something is already filed before you add a duplicate. Without list, the answer opens with the tasks that need attention (overdue, due within 7 days, high priority, in progress), each with its reason: answer from those first. Works with the app closed on a dev install (the headless daemon reads the file); an npm install has no daemon and needs the app running. Scope todo:read.

Input schema

{
  "properties": {
    "include_done": {
      "default": false,
      "description": "Include tasks already checked off. Default false: the open ones are what almost every question is actually about. Tasks moved to the ARCHIVE are never listed, only counted.",
      "type": "boolean"
    },
    "include_paths": {
      "default": false,
      "description": "Also print the full path of each attached file or folder, not only its name. Default false. A folder's content is never listed here.",
      "type": "boolean"
    },
    "limit": {
      "description": "Maximum tasks returned (default and ceiling: 300). The answer always says how many matched and how many were cut.",
      "type": "number"
    },
    "list": {
      "description": "Only this list, by displayed name (case-insensitive) or by the key shown in brackets. « The Cursor category » or « my Vitrine Studio tasks » name a list: pass it here. Omitted = every list. A name that matches nothing returns NO tasks rather than silently widening to all of them.",
      "type": "string"
    }
  },
  "type": "object"
}
sha256:0f3d0c99174958db2d5d9ea9de12eef5b3965bb6c32ee1f8caaaf980ac45381b

Output schema

Not exposed in the analyzed interface.

sha256:398b676e47d03a29016ee92fe378b8b4f1b3e965390c4c64ce78d27f79df74d1

Description hash: sha256:78ae6c7bb906deee1b943aa601765a9de5b87271d1e4014e41a4645655cd36b8

Interface hash: sha256:cbc3f77db1028ce504ddfb03962380c63ac94c5a6c0a89e3e15d16a00c966d3d

Other observed Modelara metadata
{
  "id": "mcp:mnemosyne-os/mnemosyne_todo_list",
  "title": "List To-do tasks",
  "source": {
    "type": "mcp",
    "server": "mnemosyne-os",
    "operation": "mnemosyne_todo_list"
  },
  "declared": {
    "read_only": "true",
    "destructive": "unknown",
    "idempotent": "false",
    "open_world": "unknown"
  },
  "safety": {
    "idempotent": "true",
    "retry_safe": "true"
  },
  "metadata": {
    "mcp.annotations": "present"
  },
  "fingerprint": "sha256:d113dcb1e340ff0cea77415a69cdf70e0e0f88ba99e826c5ef31f36fb84ac8ea"
}
toolread2.2.02.3.0
mnemosyne_todo_updatemnemosyne-os-mcp:tool:mnemosyne_todo_update
Observed interface and hashes

Change the user's To-do backlog: edit a task, tick it off, move it to another list, or take it out. Every operation names a task by the id from mnemosyne_todo_list, so call that first. Removing a task archives it by default, and it can be restored. Pass permanent: true only when the user asked for it to be deleted outright. The whole batch is applied in order as one save, and each operation reports its own outcome, so a stale id does not sink the ones around it. Works with the app closed on a dev install (the headless daemon reads the file); an npm install has no daemon and needs the app running. Scope todo:write.

Input schema

{
  "properties": {
    "ops": {
      "description": "The changes, applied in order. Up to 100 per call.",
      "items": {
        "properties": {
          "color": {
            "description": "edit: hex colour, or null to clear.",
            "type": [
              "string",
              "null"
            ]
          },
          "description": {
            "description": "edit: new notes. null clears them; omitting leaves them alone.",
            "type": [
              "string",
              "null"
            ]
          },
          "done": {
            "description": "complete: true ticks it, false re-opens it. Ticking records the human's streak; un-ticking never takes it back.",
            "type": "boolean"
          },
          "dropped": {
            "default": false,
            "description": "remove: archive it as « not planned » rather than as tidied away.",
            "type": "boolean"
          },
          "dueAt": {
            "description": "edit: due date as epoch milliseconds. null clears it.",
            "type": [
              "number",
              "null"
            ]
          },
          "group": {
            "description": "edit: the STEP heading this task sits under. null removes it.",
            "type": [
              "string",
              "null"
            ]
          },
          "kind": {
            "description": "edit: \"note\" turns a task into context (not ticked, counted nowhere); \"task\" turns it back.",
            "enum": [
              "note",
              "task"
            ],
            "type": "string"
          },
          "label": {
            "description": "attach: a name to show instead of the file name.",
            "type": "string"
          },
          "list": {
            "description": "move: the destination list, by displayed name or key.",
            "type": "string"
          },
          "op": {
            "description": "edit = rewrite text/description/colour/step/due date/status/priority, or turn it into a note. complete = tick or un-tick (a note cannot be ticked). move = send to another list. remove = archive it (dropped: true marks it « not planned »), or delete it with permanent: true. attach = attach a file or folder by absolute path. detach = take an attachment off by its path.",
            "enum": [
              "edit",
              "complete",
              "move",
              "remove",
              "attach",
              "detach"
            ],
            "type": "string"
          },
          "path": {
            "description": "attach / detach: the ABSOLUTE path of the file or folder. Nothing is read: only the path is stored.",
            "type": "string"
          },
          "permanent": {
            "default": false,
            "description": "remove: true DELETES the task outright, and it is not in the archive afterwards. Default false = archived, recoverable. Only pass true when the human asked to delete rather than to tidy away.",
            "type": "boolean"
          },
          "priority": {
            "description": "edit: high or low. null is normal.",
            "enum": [
              "high",
              "low",
              null
            ],
            "type": [
              "string",
              "null"
            ]
          },
          "status": {
            "description": "edit: where its work stands. null puts it back to « to do ».",
            "enum": [
              "doing",
              "waiting",
              "someday",
              null
            ],
            "type": [
              "string",
              "null"
            ]
          },
          "task": {
            "description": "The task id, exactly as mnemosyne_todo_list printed it in brackets, or its number (#42).",
            "type": "string"
          },
          "text": {
            "description": "edit: the new text. Omit to leave it alone; it can never be emptied.",
            "type": "string"
          }
        },
        "required": [
          "op",
          "task"
        ],
        "type": "object"
      },
      "type": "array"
    }
  },
  "required": [
    "ops"
  ],
  "type": "object"
}
sha256:63b95ab2be2396cba5e402e27d85cb46829740dea967c2dd585415ef0e697959

Output schema

Not exposed in the analyzed interface.

sha256:398b676e47d03a29016ee92fe378b8b4f1b3e965390c4c64ce78d27f79df74d1

Description hash: sha256:e730674f06c3402042ad1419df95e2e8430215b449ab920a31044cc76d418aff

Interface hash: sha256:52fc29b5476e0062a8c348f8e9d3ca0357b8f6254b20793cc6adad8a919aab65

Other observed Modelara metadata
{
  "id": "mcp:mnemosyne-os/mnemosyne_todo_update",
  "title": "Change the To-do backlog",
  "source": {
    "type": "mcp",
    "server": "mnemosyne-os",
    "operation": "mnemosyne_todo_update"
  },
  "declared": {
    "read_only": "false",
    "destructive": "true",
    "idempotent": "false",
    "open_world": "unknown"
  },
  "safety": {
    "idempotent": "false",
    "retry_safe": "false"
  },
  "metadata": {
    "mcp.annotations": "present"
  },
  "fingerprint": "sha256:fd5e7018bb92e94a5c9ccbdb738d89196f2eb7d34f8fd2ac97e46e8e6ee844ea"
}
tooldestructive2.2.0No change observed
mnemosyne_vault_listmnemosyne-os-mcp:tool:mnemosyne_vault_list
Observed interface and hashes

List the memory vaults this server can reach, each with its token, display name, memory count and, when the owner wrote one, a line on what it holds. Call it first when you are unsure which vault to work against, or when the user names a store you have not seen. Pass the bold token as the `vault` argument of the other tools. A vault outside this server's grant is never named here; the note under the list, when there is one, says how the user grants it.

Input schema

{
  "properties": {},
  "type": "object"
}
sha256:3178d649f32f96ea850da50c27c062f5c20cbef86dec2f5af0587dff56ebdb67

Output schema

Not exposed in the analyzed interface.

sha256:398b676e47d03a29016ee92fe378b8b4f1b3e965390c4c64ce78d27f79df74d1

Description hash: sha256:0ddc5dfdb898d79b46254346d853e8f2229e89d4bbaf1754cd9b7584e811ac44

Interface hash: sha256:0c7de9c1942340f0b2719b4c9277b55f718f93f03471eef67e02aac653fa9f20

Other observed Modelara metadata
{
  "id": "mcp:mnemosyne-os/mnemosyne_vault_list",
  "title": "List vaults",
  "source": {
    "type": "mcp",
    "server": "mnemosyne-os",
    "operation": "mnemosyne_vault_list"
  },
  "declared": {
    "read_only": "true",
    "destructive": "unknown",
    "idempotent": "false",
    "open_world": "unknown"
  },
  "safety": {
    "idempotent": "true",
    "retry_safe": "true"
  },
  "metadata": {
    "mcp.annotations": "present"
  },
  "fingerprint": "sha256:96a6a3132c3da2d7d3b749ed3797aaa056de2023c75ed8d5f6224b393ea275fc"
}
toolread2.2.02.4.0

STATIC COMPARISONS

Release history

2 release transitions
  1. 2 added0 removed2 schema0 write1 destructive0 unknown

    Nov 16, 2026· Pending human review

  2. 4 added0 removed2 schema3 write0 destructive0 unknown

    Nov 16, 2026· Pending human review