developer-tools Static interface comparison

Mnemosyne OS

2.2.0 2.3.0

FAILPending human review

3 new write capabilities require explicit review.

View the upgrade review as Markdown

A shareable summary of this comparison for an issue, pull request or ticket. Select the text and copy it.

Tools25 → 29
Added4
Removed0
Schema changes2
Input schema changes2
Output schema changes0
Unclassified schema changes0
Description changes2
New write authority3
New destructive authority0
New unknown authority0
AnalysisFAIL

INTERFACE DIFF

What changed, tool by tool

Classification from the recorded findings · definitions from the published snapshots

6 tools with recorded changes

  1. mnemosyne_dream_bridgesModifiedSchemaDescriptionREVIEWINFO

    What changed

    • mnemosyne_dream_bridges — Schema change: input property added: include_unresolved.
    • mnemosyne_dream_bridges: description changed.

    Why it could matter

    • Calls or consumers built for the earlier schema may need to change.
    • Models choose and call tools partly from their descriptions.

    What to verify

    • Validate representative existing tool calls against the new argument schema.
    • Spot-check that agents still select this tool as intended.
    Declared effect
    read → read
    Description
    PreviousList 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. 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.
    CurrentList 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

    PropertyPrevious (2.2.0)Current (2.3.0)Change
    include_unresolvednot presentbooleanoptional
    • Added
    5 unchanged entries
    • (schema root)
    • chronicle_id
    • limit
    • min_dbs
    • session_id

    No output schema is recorded for this tool in either release.

    Exact JSON from the published snapshots

    Loads the exact recorded definitions of this tool from both releases. The full snapshots are linked under Evidence and provenance.

    Link to this tool
  2. mnemosyne_review_getAddedAddedINFO

    What changed

    • New read-only tool mnemosyne_review_get.

    Why it could matter

    • The release adds a capability without declared write authority.

    What to verify

    • Decide whether agents should be offered this tool.
    Declared effect
    — → read

    Input schema

    PropertyPrevious (2.2.0)Current (2.3.0)Change
    (schema root)not presentobject
    • Added
    idnot presentstringrequired
    • Added

    No output schema is recorded for this tool in either release.

    Exact JSON from the published snapshots

    Loads the exact recorded definitions of this tool from both releases. The full snapshots are linked under Evidence and provenance.

    Link to this tool
  3. mnemosyne_review_openAddedAuthorityFAIL

    What changed

    • New tool mnemosyne_review_open, declared with write effect.

    Why it could matter

    • Agents connected to this release can be offered an operation that changes data, which the earlier release did not expose.

    What to verify

    • Review whether this operation should be exposed through your agent’s tool configuration.
    Declared effect
    — → write

    Input schema

    PropertyPrevious (2.2.0)Current (2.3.0)Change
    (schema root)not presentobject
    • Added
    fpsnot presentnumberoptional
    • Added
    parentIdnot presentstringoptional
    • Added
    pathnot presentstringrequired
    • Added
    projectnot presentstringoptional
    • Added
    questionnot presentstringoptional
    • Added
    titlenot presentstringrequired
    • Added

    No output schema is recorded for this tool in either release.

    Exact JSON from the published snapshots

    Loads the exact recorded definitions of this tool from both releases. The full snapshots are linked under Evidence and provenance.

    Link to this tool
  4. mnemosyne_review_waitAddedAuthorityFAIL

    What changed

    • New tool mnemosyne_review_wait, declared with write effect.

    Why it could matter

    • Agents connected to this release can be offered an operation that changes data, which the earlier release did not expose.

    What to verify

    • Review whether this operation should be exposed through your agent’s tool configuration.
    Declared effect
    — → write

    Input schema

    PropertyPrevious (2.2.0)Current (2.3.0)Change
    (schema root)not presentobject
    • Added
    afternot presentnumberoptional
    • Added
    idnot presentstringrequired
    • Added
    timeout_secondsnot presentnumberoptional
    • Added

    No output schema is recorded for this tool in either release.

    Exact JSON from the published snapshots

    Loads the exact recorded definitions of this tool from both releases. The full snapshots are linked under Evidence and provenance.

    Link to this tool
  5. mnemosyne_telegram_sendAddedAuthorityFAIL

    What changed

    • New tool mnemosyne_telegram_send, declared with write effect.

    Why it could matter

    • Agents connected to this release can be offered an operation that changes data, which the earlier release did not expose.

    What to verify

    • Review whether this operation should be exposed through your agent’s tool configuration.
    Declared effect
    — → write

    Input schema

    PropertyPrevious (2.2.0)Current (2.3.0)Change
    (schema root)not presentobject
    • Added
    captionnot presentstringoptional
    • Added
    pathnot presentstringrequired
    • Added

    No output schema is recorded for this tool in either release.

    Exact JSON from the published snapshots

    Loads the exact recorded definitions of this tool from both releases. The full snapshots are linked under Evidence and provenance.

    Link to this tool
  6. mnemosyne_todo_listModifiedSchemaDescriptionREVIEWINFO

    What changed

    • mnemosyne_todo_list — Schema change: input property list changed: description "Only this list, by… → "Only this list, by….
    • mnemosyne_todo_list: description changed.

    Why it could matter

    • Calls or consumers built for the earlier schema may need to change.
    • Models choose and call tools partly from their descriptions.

    What to verify

    • Validate representative existing tool calls against the new argument schema.
    • Spot-check that agents still select this tool as intended.
    Declared effect
    read → read
    Description
    PreviousRead the user's To-do backlog back: the lists that exist 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. 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.
    CurrentRead 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

    PropertyPrevious (2.2.0)Current (2.3.0)Change
    liststringoptionalstringoptional
    • Description changed
    4 unchanged entries
    • (schema root)
    • include_done
    • include_paths
    • limit

    No output schema is recorded for this tool in either release.

    Exact JSON from the published snapshots

    Loads the exact recorded definitions of this tool from both releases. The full snapshots are linked under Evidence and provenance.

    Link to this tool
23 tools with no recorded change
  • mnemosyne_about
  • mnemosyne_agenda_add
  • mnemosyne_agenda_list
  • mnemosyne_agenda_remove
  • mnemosyne_agenda_update
  • mnemosyne_agent_collisions
  • mnemosyne_agent_files
  • mnemosyne_agent_list
  • mnemosyne_cockpit_update
  • mnemosyne_git_log
  • mnemosyne_memory_ask
  • mnemosyne_memory_ingest
  • mnemosyne_memory_query
  • mnemosyne_pheme_radar
  • mnemosyne_pheme_watch
  • mnemosyne_position_get
  • mnemosyne_position_update
  • mnemosyne_resonance_list
  • mnemosyne_spine_assignments
  • mnemosyne_todo_add
  • mnemosyne_todo_categories
  • mnemosyne_todo_update
  • mnemosyne_vault_list

Evidence and provenance

Static findings
FAIL from Modelara’s automated analysis of the two declared interfaces. A policy outcome, not a security rating or a runtime test.
Human review
Pending human review. No reviewer decision is recorded for this comparison.
Published evidence
Part of Valyd’s published data generation listed in valyd-data-source.json. Published records are not rewritten.
Show provenance and exact files
Comparison
mnemosyne-os-mcp/2.2.0~2.3.0The published comparison identifier.
Analyzer
modelara v0.1.8Policy version not recorded.Analyzed Nov 16, 2026, 10:06 PM UTC.
Distribution
npm
Previous release
@mnemosyne_os/mcp@2.2.025 tools · published by the registry Oct 29, 2026, 5:43 PM UTCArtifact integritysha512-vQ956Udkzwcmc3ze7giAw8kkLat/z8gqzhCjGBzrzA3WYwMEKefoQJUZyprJEjPcoCKVvNFON21irSuldWkebQ==Interface fingerprintsha256:ccc62c3d0f45c477d6f111a34da4e785e29202d78b92ffe4f6c933ae045609acExact interface snapshot (JSON)
Current release
@mnemosyne_os/mcp@2.3.029 tools · published by the registry Nov 9, 2026, 7:14 AM UTCArtifact integritysha512-8GfrVBbZNUamhqYrzg6Im2Wb6bIXrsC+J0jZLwNwqv60iGDwOHO/2J+CD+JvY7VX58YrM5HYGVhWVl+r588Oiw==Interface fingerprintsha256:c8b47f9e59b2cac92f99789f5fdcdf492e24472fef05fb81b9d1aa342845c093Exact interface snapshot (JSON)

The integrity value or digest identifies the exact artifact that was inspected; it does not show that the package is safe. The interface fingerprint identifies the declared tools and schemas Modelara recorded for that release.

MODELARA ANALYSIS OUTPUT

Interface findings

Analyzed with Modelara v0.1.8 · Nov 16, 2026

Policy failures

3 findings
  • mnemosyne_review_openFAILAdded

    New capability with write effect

    Authority
    —WRITE
  • mnemosyne_review_waitFAILAdded

    New capability with write effect

    Authority
    —WRITE
  • mnemosyne_telegram_sendFAILAdded

    New capability with write effect

    Authority
    —WRITE

Requires review

2 findings
  • mnemosyne_dream_bridgesREVIEWChanged

    Input schema changed

    Show details
    • input property added: include_unresolved
    Authority
    READREAD
    Authority unchanged
  • mnemosyne_todo_listREVIEWChanged

    Input schema changed

    Show details
    • input property list changed: description "Only this list, by… → "Only this list, by…
    Authority
    READREAD
    Authority unchanged

Informational

3 findings
  • mnemosyne_review_getINFOAdded

    New read-only capability

    Authority
    —READ
  • mnemosyne_dream_bridgesINFOChanged

    Description changed

    Authority
    READREAD
    Authority unchanged
  • mnemosyne_todo_listINFOChanged

    Description changed

    Authority
    READREAD
    Authority unchanged
What do these results mean?
PASS
No policy-significant interface change detected.
REVIEW
Interface changes require inspection.
FAIL
A policy-significant change was detected.
INVALID
The comparison could not be evaluated reliably.
← Back to Mnemosyne OS release history