developer-tools Static interface comparison
Fantasy Football Manager
0.3.2 0.3.3
4 new write capabilities require explicit review.4 new destructive 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.
INTERFACE DIFF
What changed, tool by tool
- 0Added tools
- 0Removed tools
- 17Modified tools
- 9Input-schema changes
- 0Output-schema changes
- 8Declared authority changes
17 tools with recorded changes
execute_demo_actionModifiedWhat changed
- execute_demo_action: declared effect changed from write to destructive.
- execute_demo_action — Schema change: input property confirmation changed: +description="Set true only afte….
- execute_demo_action — Schema change: input property proposal_id changed: +description="Exact proposal_id ….
- execute_demo_action: description changed.
- execute_demo_action: declared metadata changed (declared idempotent: false → true).
Why it could matter
- An existing tool name now declares stronger authority, so a configuration that allowed it before may allow more now.
- Calls or consumers built for the earlier schema may need to change.
- Models choose and call tools partly from their descriptions.
- Annotations such as idempotent or open-world hints can inform how clients treat a tool.
What to verify
- Re-confirm whether agents should keep access to this tool.
- Validate representative existing tool calls against the new argument schema.
- Spot-check that agents still select this tool as intended.
- Confirm any client logic that relies on these annotations.
- Declared effect
- write → write
- Description
- PreviousExecute a saved synthetic proposal. Set confirmation only after the user approves that proposal in review mode.
- CurrentExecute an existing proposal against synthetic demo state only. First call prepare_action, then obtain approval for that exact proposal when requires_confirmation is true. Execution rechecks snapshot and configuration revisions, freshness, automation modes, and action limits before changing state. Returns status='executed', scope, action, proposal_id, and the new revision, and records one audit event. Retrying a completed proposal returns its original result with idempotent_replay=true, without another state change or approval check. Unknown or changed proposals fail. Use the ESPN companion for live submissions.
Input schema
Property Previous (0.3.2) Current (0.3.3) Change confirmationboolean, default=falseoptional boolean, default=falseoptional - Description changed
proposal_idstringrequired stringrequired - Description changed
1 unchanged entry
(schema root)
Output schema
Output schema unchanged.
1 unchanged entry
(schema root)
Link to this toolExact 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.
get_action_historyModifiedWhat changed
- get_action_history — Schema change: input property limit changed: +description="Maximum audit reco….
- get_action_history: 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 local state changes and completed demo actions from the audit history.
- CurrentRead recent audit records from this data directory without changing state. Returns history entries with id, at, event, and detail, ordered from newest to oldest. Records include snapshot imports, configuration changes, prepared proposals, and completed demo actions. An action_prepared event records preparation, not execution. Calculation evidence is separate from this audit history. Works before a snapshot is loaded.
Input schema
Property Previous (0.3.2) Current (0.3.3) Change limitinteger, default=50optional integer, default=50optional - Description changed
1 unchanged entry
(schema root)
Output schema
Output schema unchanged.
1 unchanged entry
(schema root)
Link to this toolExact 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.
get_capabilitiesModifiedWhat changed
- get_capabilities: description changed.
Why it could matter
- Models choose and call tools partly from their descriptions.
What to verify
- Spot-check that agents still select this tool as intended.
- Declared effect
- read → read
- Description
- PreviousRead supported operations and the effective execution scope.
- CurrentRead this server's capabilities before selecting analysis or execution tools. Returns supported actions, effective automation modes, pause state, snapshot presence, and state and configuration revisions. This server analyzes imported data and executes synthetic demo actions only. The response separately describes the ESPN companion for live observations, draft picks, and lineup changes. Works before a snapshot is loaded and does not change saved state.
Input schema
Input schema unchanged.
1 unchanged entry
(schema root)
Output schema
Output schema unchanged.
1 unchanged entry
(schema root)
Link to this toolExact 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.
get_draft_recommendationsModifiedWhat changed
- get_draft_recommendations: description changed.
Why it could matter
- Models choose and call tools partly from their descriptions.
What to verify
- Spot-check that agents still select this tool as intended.
- Declared effect
- read → read
- Description
- PreviousRead current monitor results. Changed or stale snapshots suppress cached results.
- CurrentRead cached results and status from the local draft monitor without running a simulation. Returns running, status, error, current, trial counters, and latest when the cached inputs remain usable. latest is null when revisions change, the snapshot becomes stale or incomplete, automation pauses, or no result exists. Completed trials include work across the monitor session, while current_state_trials counts only the current inputs. Use start_draft_monitor to produce these results or recommend_draft for an independent single batch. This tool does not fetch observations or append calculation evidence.
Input schema
Input schema unchanged.
1 unchanged entry
(schema root)
Output schema
Output schema unchanged.
1 unchanged entry
(schema root)
Link to this toolExact 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.
get_manager_configModifiedWhat changed
- get_manager_config: description changed.
Why it could matter
- Models choose and call tools partly from their descriptions.
What to verify
- Spot-check that agents still select this tool as intended.
- Declared effect
- read → read
- Description
- PreviousRead user strategy, automation modes, limits, and config revision.
- CurrentRead the full current configuration and its config_revision. Returns strategies, automation presets and per-action modes, pause state, and user limits. Works before a snapshot is loaded and does not change saved state. Use this response as the starting configuration for update_manager_config to preserve existing settings.
Input schema
Input schema unchanged.
1 unchanged entry
(schema root)
Output schema
Output schema unchanged.
1 unchanged entry
(schema root)
Link to this toolExact 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.
get_power_rankingsModifiedWhat changed
- get_power_rankings: declared effect changed from read to write.
- get_power_rankings: description changed.
- get_power_rankings: declared metadata changed (declared idempotent: true → false).
Why it could matter
- An existing tool name now declares stronger authority, so a configuration that allowed it before may allow more now.
- Models choose and call tools partly from their descriptions.
- Annotations such as idempotent or open-world hints can inform how clients treat a tool.
What to verify
- Re-confirm whether agents should keep access to this tool.
- Spot-check that agents still select this tool as intended.
- Confirm any client logic that relies on these annotations.
- Declared effect
- read → write
- Description
- PreviousCompare each team's best legal weekly lineup. These are not championship odds.
- CurrentCompare every team's best legal weekly lineup by projected points. Requires current season data, weekly inputs for every team, and verified locks with league scope. Returns rankings, source details, warnings, and input revisions with status='ok' or status='incomplete'. Incomplete team lineups produce null ranks, while missing league lock coverage produces an empty rankings list. These estimates describe one week and are not championship probabilities. Appends calculation evidence without changing rosters. Use recommend_lineup to analyze only the selected team.
Input schema
Input schema unchanged.
1 unchanged entry
(schema root)
Output schema
Output schema unchanged.
1 unchanged entry
(schema root)
Link to this toolExact 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.
get_source_statusModifiedWhat changed
- get_source_status: description changed.
Why it could matter
- Models choose and call tools partly from their descriptions.
What to verify
- Spot-check that agents still select this tool as intended.
- Declared effect
- read → read
- Description
- PreviousRead observation age, source completeness, and monitor status.
- CurrentCheck the freshness and completeness of the saved observation. Requires an imported snapshot or loaded demo. Returns source fields, age_seconds, the phase-specific maximum_age_seconds, stale, and local draft monitor status. This tool reads saved data and does not refresh the source or run a simulation. Use the ESPN companion for new browser observations or import_league_snapshot for replacement data.
Input schema
Input schema unchanged.
1 unchanged entry
(schema root)
Output schema
Output schema unchanged.
1 unchanged entry
(schema root)
Link to this toolExact 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.
get_teamModifiedWhat changed
- get_team: description changed.
Why it could matter
- Models choose and call tools partly from their descriptions.
What to verify
- Spot-check that agents still select this tool as intended.
- Declared effect
- read → read
- Description
- PreviousRead the selected team and its players from the current snapshot.
- CurrentRead the selected team's roster from the saved snapshot. Requires an imported snapshot or loaded demo. Returns the snapshot revision, league context, roster and reserve players, lineup, league rules, and budget when supplied. Use recommend_lineup to calculate a lineup or get_source_status to check observation freshness. This tool does not fetch new data or change saved state.
Input schema
Input schema unchanged.
1 unchanged entry
(schema root)
Output schema
Output schema unchanged.
1 unchanged entry
(schema root)
Link to this toolExact 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.
import_league_snapshotModifiedWhat changed
- import_league_snapshot: declared effect changed from write to destructive.
- import_league_snapshot — Schema change: input property expected_revision changed: +description="Current snapshot r….
- import_league_snapshot — Schema change: input property snapshot changed: additionalProperties true → false, +description="Complete league ob…, +properties={"budget":{"anyOf":…, +required=["league_id","team_…, title "Snapshot" → "LeagueSnapshot".
- import_league_snapshot: description changed.
Why it could matter
- An existing tool name now declares stronger authority, so a configuration that allowed it before may allow more now.
- Calls or consumers built for the earlier schema may need to change.
- Models choose and call tools partly from their descriptions.
What to verify
- Re-confirm whether agents should keep access to this tool.
- Validate representative existing tool calls against the new argument schema.
- Spot-check that agents still select this tool as intended.
- Declared effect
- write → write
- Description
- PreviousImport a complete JSON snapshot. Supply the current revision when replacing state.
- CurrentValidate and save a complete league snapshot supplied by the caller. Use the ESPN companion to fetch live observations or load_demo for fictional data. Replacement requires the current snapshot revision and must preserve observation order, confirmed draft history, and pending browser action context. Changing the league, selected team, or season resets configuration to defaults. Records the import, increments the snapshot revision, and invalidates unexecuted proposals and monitor results. Returns status='imported', revision, config_revision, and config_reset, or an error without replacing state.
Input schema
Property Previous (0.3.2) Current (0.3.3) Change expected_revisionanyOf (2), default=nulloptional anyOf (2), default=nulloptional - Description changed
snapshotobject, additionalProperties=truerequired object, additionalProperties=falserequired - Constraint: additionalProperties
- Description changed
- Title changed
additionalProperties: true → falsesnapshot › budgetnot present anyOf (2), default=nulloptional - Added
snapshot › league_idnot present string, maxLength=120, minLength=1required - Added
snapshot › phasenot present string, default="season", enum (2)optional - Added
snapshot › picksnot present arrayoptional - Added
snapshot › picks › [items]not present object, additionalProperties=false - Added
snapshot › picks › [items] › pick_nonot present integer, maximum=1280, minimum=1required - Added
snapshot › picks › [items] › player_idnot present stringrequired - Added
snapshot › picks › [items] › slotnot present integer, maximum=32, minimum=1required - Added
snapshot › playersnot present array, maxItems=10000, minItems=1required - Added
snapshot › players › [items]not present object, additionalProperties=false - Added
snapshot › players › [items] › adpnot present number, default=999, exclusiveMinimum=0optional - Added
snapshot › players › [items] › availabilitynot present string, default="UNKNOWN"optional - Added
snapshot › players › [items] › byenot present anyOf (2), default=nulloptional - Added
snapshot › players › [items] › eligible_positionsnot present arrayoptional - Added
snapshot › players › [items] › eligible_positions › [items]not present string - Added
snapshot › players › [items] › idnot present string, maxLength=120, minLength=1required - Added
snapshot › players › [items] › lockednot present boolean, default=falseoptional - Added
snapshot › players › [items] › namenot present string, maxLength=160, minLength=1required - Added
snapshot › players › [items] › positionnot present string, enum (6)required - Added
snapshot › players › [items] › projectionnot present anyOf (2), default=nulloptional - Added
snapshot › players › [items] › teamnot present string, default=""optional - Added
snapshot › players › [items] › weekly_ceilingnot present anyOf (2), default=nulloptional - Added
snapshot › players › [items] › weekly_floornot present anyOf (2), default=nulloptional - Added
snapshot › players › [items] › weekly_projectionnot present anyOf (2), default=nulloptional - Added
snapshot › rulesnot present object, additionalProperties=falserequired - Added
snapshot › rules › benchnot present integer, default=5, maximum=30, minimum=0optional - Added
snapshot › rules › capsnot present object, additionalProperties {…}optional - Added
snapshot › rules › flex_eligiblenot present arrayoptional - Added
snapshot › rules › flex_eligible › [items]not present string - Added
snapshot › rules › irnot present integer, default=2, maximum=10, minimum=0optional - Added
snapshot › rules › roundsnot present integer, default=14, maximum=40, minimum=1optional - Added
snapshot › rules › slotnot present integer, default=1, maximum=32, minimum=1optional - Added
snapshot › rules › snakenot present boolean, default=trueoptional - Added
snapshot › rules › startersnot present object, additionalProperties {…}optional - Added
snapshot › rules › teamsnot present integer, default=14, maximum=32, minimum=2optional - Added
snapshot › schema_versionnot present integer, const=1, default=1optional - Added
snapshot › seasonnot present integer, maximum=2100, minimum=2020required - Added
snapshot › sourcenot present object, additionalProperties=falserequired - Added
snapshot › source › browsernot present anyOf (2), default=nulloptional - Added
snapshot › source › completenot present boolean, default=trueoptional - Added
snapshot › source › locks_scopenot present string, enum (2)optional - Added
snapshot › source › locks_verifiednot present boolean, default=falseoptional - Added
snapshot › source › notesnot present arrayoptional - Added
snapshot › source › notes › [items]not present string - Added
snapshot › source › observed_atnot present string, format="date-time"required - Added
snapshot › source › projections_observed_atnot present anyOf (2), default=nulloptional - Added
snapshot › source › providernot present string, maxLength=80, minLength=1required - Added
snapshot › source › syntheticnot present boolean, default=falseoptional - Added
snapshot › team_idnot present string, maxLength=120, minLength=1required - Added
snapshot › teamsnot present array, maxItems=32, minItems=2required - Added
snapshot › teams › [items]not present object, additionalProperties=false - Added
snapshot › teams › [items] › idnot present string, maxLength=120, minLength=1required - Added
snapshot › teams › [items] › lineupnot present object, additionalProperties {…}optional - Added
snapshot › teams › [items] › namenot present string, maxLength=160, minLength=1required - Added
snapshot › teams › [items] › reserve_idsnot present arrayoptional - Added
snapshot › teams › [items] › reserve_ids › [items]not present string - Added
snapshot › teams › [items] › roster_idsnot present arrayoptional - Added
snapshot › teams › [items] › roster_ids › [items]not present string - Added
snapshot › teams › [items] › slotnot present integer, maximum=32, minimum=1required - Added
snapshot › weeknot present integer, default=1, maximum=18, minimum=1optional - Added
1 unchanged entry
(schema root)
Output schema
Output schema unchanged.
1 unchanged entry
(schema root)
Link to this toolExact 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.
load_demoModifiedWhat changed
- load_demo: declared effect changed from write to destructive.
- load_demo — Schema change: input property mode changed: +description="Fictional scenario…, +enum=["season","draft"].
- load_demo: description changed.
Why it could matter
- An existing tool name now declares stronger authority, so a configuration that allowed it before may allow more now.
- Calls or consumers built for the earlier schema may need to change.
- Models choose and call tools partly from their descriptions.
What to verify
- Re-confirm whether agents should keep access to this tool.
- Validate representative existing tool calls against the new argument schema.
- Spot-check that agents still select this tool as intended.
- Declared effect
- write → write
- Description
- PreviousLoad fictional data. This replaces local state and resets config when the league changes.
- CurrentLoad a fictional scenario for local testing or demonstrations. Use import_league_snapshot for your own data. Requires an empty data directory or existing synthetic state and preserves confirmed draft history checks. Stops the local draft monitor, replaces the snapshot, and resets configuration when the league changes. Records the import and returns status='imported', revision, config_revision, and config_reset. This tool does not connect to ESPN.
Input schema
Property Previous (0.3.2) Current (0.3.3) Change modestring, default="season"optional string, default="season", enum (2)optional - Description changed
- Constraint: enum
enum: — → ["season","draft"]1 unchanged entry
(schema root)
Output schema
Output schema unchanged.
1 unchanged entry
(schema root)
Link to this toolExact 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.
prepare_actionModifiedWhat changed
- prepare_action — Schema change: input property action changed: +description="Synthetic action: …, +enum=["draft_pick","set_….
- prepare_action — Schema change: input property payload changed: -additionalProperties, +anyOf=[{"additionalProper…, +description="Exact action data.…, +examples=[{"player_id":"demo…, -type.
- prepare_action: 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
- write → write
- Description
- PreviousCheck an exact synthetic action and save a revision-bound proposal. This does not execute it.
- CurrentValidate a synthetic action and save a proposal without executing it. Requires fresh, complete synthetic data, unpaused automation, and review or automatic mode for every action component. Checks action-specific player eligibility, locks, projections, roster rules, and user limits before saving the proposal. Returns proposal_id, normalized payload, mode, requires_confirmation, scope, and the bound snapshot and configuration revisions. Each call creates a new proposal and audit record. Use execute_demo_action with that exact proposal_id for execution. Use the ESPN companion to prepare live draft picks or lineup changes.
Input schema
Property Previous (0.3.2) Current (0.3.3) Change actionstringrequired string, enum (5)required - Description changed
- Constraint: enum
enum: — → ["draft_pick","set_lineup","waiver_claim","free_agent_add","drop_player"]payloadobject, additionalProperties=truerequired anyOf (3), examples (1)required - Constraint: additionalProperties
- Changed: anyOf
- Description changed
- Changed: examples
- Type changed
additionalProperties: true → —anyOf: — → [{"additionalProperties":false,"description":"Payload for draft_pick or drop_player.","properties":{"player_id":{"description":"Exact player ID from the saved snapshot.","title":"Player Id","type":"string"}},"required":["player_id"],"title"… (truncated: full value in the exact JSON)examples: — → [{"player_id":"demo-rb-001"}]type: "object" → —1 unchanged entry
(schema root)
Output schema
Output schema unchanged.
1 unchanged entry
(schema root)
Link to this toolExact 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.
rank_waiver_candidatesModifiedWhat changed
- rank_waiver_candidates: declared effect changed from read to write.
- rank_waiver_candidates — Schema change: input property limit changed: +description="Maximum number of ….
- rank_waiver_candidates: description changed.
- rank_waiver_candidates: declared metadata changed (declared idempotent: true → false).
Why it could matter
- An existing tool name now declares stronger authority, so a configuration that allowed it before may allow more now.
- Calls or consumers built for the earlier schema may need to change.
- Models choose and call tools partly from their descriptions.
- Annotations such as idempotent or open-world hints can inform how clients treat a tool.
What to verify
- Re-confirm whether agents should keep access to this tool.
- Validate representative existing tool calls against the new argument schema.
- Spot-check that agents still select this tool as intended.
- Confirm any client logic that relies on these annotations.
- Declared effect
- read → write
- Description
- PreviousRank alternative acquisitions using weekly projections and configured limits.
- CurrentRank alternative single-player acquisitions using the saved weekly snapshot and user limits. Requires current season data, verified locks, weekly projections, and known budget and pending commitments. Returns recommendations with add and optional drop IDs, lineup improvement, and maximum FAAB bids, plus revisions and input warnings. Status is ok, incomplete for missing inputs, or blocked when the weekly move limit is reached. Unrostered players are not verified free agents, and this tool does not estimate winning bids or submit claims. Appends calculation evidence. Use prepare_action only for synthetic acquisitions that pass action checks.
Input schema
Property Previous (0.3.2) Current (0.3.3) Change limitinteger, default=10optional integer, default=10optional - Description changed
1 unchanged entry
(schema root)
Output schema
Output schema unchanged.
1 unchanged entry
(schema root)
Link to this toolExact 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.
recommend_draftModifiedWhat changed
- recommend_draft: declared effect changed from read to write.
- recommend_draft — Schema change: input property seed changed: +description="Random seed for th….
- recommend_draft — Schema change: input property trials changed: +description="Simulation trials ….
- recommend_draft: description changed.
- recommend_draft: declared metadata changed (declared idempotent: true → false).
Why it could matter
- An existing tool name now declares stronger authority, so a configuration that allowed it before may allow more now.
- Calls or consumers built for the earlier schema may need to change.
- Models choose and call tools partly from their descriptions.
- Annotations such as idempotent or open-world hints can inform how clients treat a tool.
What to verify
- Re-confirm whether agents should keep access to this tool.
- Validate representative existing tool calls against the new argument schema.
- Spot-check that agents still select this tool as intended.
- Confirm any client logic that relies on these annotations.
- Declared effect
- read → write
- Description
- PreviousSimulate draft choices from the current snapshot. Results are estimates.
- CurrentRun one bounded draft simulation against the current imported draft snapshot. Use start_draft_monitor for repeated batches or get_draft_recommendations to read that monitor's cached results. Requires draft data with consistent pick history and full-season projections for available and selected-team players. Returns estimated recommendations, trial counts, warnings, revisions, and status such as ready, stale_snapshot, incomplete_snapshot, or roster_complete. current checks revision and snapshot freshness after calculation, but does not grant action permission. Appends a calculation evidence record, including discarded work, without changing rosters or submitting picks.
Input schema
Property Previous (0.3.2) Current (0.3.3) Change seedinteger, default=1optional integer, default=1optional - Description changed
trialsinteger, default=40optional integer, default=40optional - Description changed
1 unchanged entry
(schema root)
Output schema
Output schema unchanged.
1 unchanged entry
(schema root)
Link to this toolExact 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.
recommend_lineupModifiedWhat changed
- recommend_lineup: declared effect changed from read to write.
- recommend_lineup: description changed.
- recommend_lineup: declared metadata changed (declared idempotent: true → false).
Why it could matter
- An existing tool name now declares stronger authority, so a configuration that allowed it before may allow more now.
- Models choose and call tools partly from their descriptions.
- Annotations such as idempotent or open-world hints can inform how clients treat a tool.
What to verify
- Re-confirm whether agents should keep access to this tool.
- Spot-check that agents still select this tool as intended.
- Confirm any client logic that relies on these annotations.
- Declared effect
- read → write
- Description
- PreviousFind the best legal weekly lineup for the selected strategy.
- CurrentCalculate the selected team's best legal weekly lineup for the configured season strategy. Uses the saved season snapshot, weekly projections, verified player locks, and configured source age limits. Returns status='ok' with lineup, projected points, and improvement, or status='incomplete' with errors and missing-input details. Locked players remain in their current slots, including unavailable players whose games have locked. Appends calculation evidence without changing the lineup, and includes the input snapshot and configuration revisions. Use prepare_action for a synthetic proposal or the ESPN companion for live lineup changes.
Input schema
Input schema unchanged.
1 unchanged entry
(schema root)
Output schema
Output schema unchanged.
1 unchanged entry
(schema root)
Link to this toolExact 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.
start_draft_monitorModifiedWhat changed
- start_draft_monitor — Schema change: input property interval_seconds changed: +description="Pause in seconds a….
- start_draft_monitor — Schema change: input property trials changed: +description="Trials per simulat….
- start_draft_monitor: 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
- write → write
- Description
- PreviousStart repeated local simulation batches. Import new observations to keep results current.
- CurrentStart background draft simulations and return the initial monitor status. Requires a loaded draft snapshot and no active local draft monitor. Each batch uses the latest imported snapshot and appends calculation evidence, including discarded batches. The monitor waits while paused or while the snapshot is stale or incomplete, and clears obsolete cached results. Import new observations to maintain freshness. This tool neither reads a browser nor submits picks. Use get_draft_recommendations for results, stop_draft_monitor to stop, or recommend_draft for one synchronous calculation.
Input schema
Property Previous (0.3.2) Current (0.3.3) Change interval_secondsnumber, default=2optional number, default=2optional - Description changed
trialsinteger, default=40optional integer, default=40optional - Description changed
1 unchanged entry
(schema root)
Output schema
Output schema unchanged.
1 unchanged entry
(schema root)
Link to this toolExact 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.
stop_draft_monitorModifiedWhat changed
- stop_draft_monitor: description changed.
- stop_draft_monitor: declared metadata changed (declared idempotent: false → true).
Why it could matter
- Models choose and call tools partly from their descriptions.
- Annotations such as idempotent or open-world hints can inform how clients treat a tool.
What to verify
- Spot-check that agents still select this tool as intended.
- Confirm any client logic that relies on these annotations.
- Declared effect
- write → write
- Description
- PreviousStop the monitor after its current bounded batch.
- CurrentRequest shutdown of the local draft monitor and wait up to five seconds. Returns monitor status and counters, with running=true and status='stopping' if the active batch still needs to finish. Use get_draft_recommendations to confirm running=false before starting another monitor. Repeated calls are safe when the monitor is already stopped. This tool does not stop the ESPN companion or change saved automation configuration.
Input schema
Input schema unchanged.
1 unchanged entry
(schema root)
Output schema
Output schema unchanged.
1 unchanged entry
(schema root)
Link to this toolExact 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.
update_manager_configModifiedWhat changed
- update_manager_config: declared effect changed from write to destructive.
- update_manager_config — Schema change: input property config changed: additionalProperties true → false, +description="Replacement config…, +properties={"automation":{"add…, title "Config" → "ManagerConfig".
- update_manager_config — Schema change: input property expected_revision changed: +description="Current config_rev….
- update_manager_config: description changed.
Why it could matter
- An existing tool name now declares stronger authority, so a configuration that allowed it before may allow more now.
- Calls or consumers built for the earlier schema may need to change.
- Models choose and call tools partly from their descriptions.
What to verify
- Re-confirm whether agents should keep access to this tool.
- Validate representative existing tool calls against the new argument schema.
- Spot-check that agents still select this tool as intended.
- Declared effect
- write → write
- Description
- PreviousReplace the full config using its current revision. This invalidates prepared actions.
- CurrentReplace the saved configuration using its current configuration revision. Read get_manager_config first to preserve settings that you do not intend to change. Omitted fields use model defaults rather than existing values, including when config is an empty object. Validates strategies and limits, records the change, and invalidates unexecuted proposals and monitor results. Returns status='updated' and the new config_revision, or a revision conflict without changing configuration. This tool does not execute actions or start a monitor.
Input schema
Property Previous (0.3.2) Current (0.3.3) Change configobject, additionalProperties=truerequired object, additionalProperties=falserequired - Constraint: additionalProperties
- Description changed
- Title changed
additionalProperties: true → falseconfig › automationnot present object, additionalProperties=falseoptional - Added
config › automation › actionsnot present object, additionalProperties {…}optional - Added
config › automation › pausednot present boolean, default=falseoptional - Added
config › automation › presetnot present string, default="advisory", enum (4)optional - Added
config › limitsnot present object, additionalProperties=falseoptional - Added
config › limits › allowed_drop_idsnot present arrayoptional - Added
config › limits › allowed_drop_ids › [items]not present string - Added
config › limits › batch_trialsnot present integer, default=40, maximum=500, minimum=1optional - Added
config › limits › drop_modenot present string, default="listed_only", enum (2)optional - Added
config › limits › faab_per_claimnot present integer, default=10, minimum=0optional - Added
config › limits › faab_per_seasonnot present integer, default=70, minimum=0optional - Added
config › limits › faab_per_weeknot present integer, default=20, minimum=0optional - Added
config › limits › faab_reservenot present integer, default=30, minimum=0optional - Added
config › limits › max_adp_reachnot present anyOf (2), default=nulloptional - Added
config › limits › max_draft_age_secondsnot present integer, default=15, maximum=300, minimum=1optional - Added
config › limits › max_projection_age_secondsnot present integer, default=3600, maximum=604800, minimum=1optional - Added
config › limits › max_season_age_secondsnot present integer, default=300, maximum=86400, minimum=1optional - Added
config › limits › max_weekly_movesnot present integer, default=3, maximum=100, minimum=0optional - Added
config › limits › min_lineup_improvementnot present number, default=1.5, minimum=0optional - Added
config › limits › protected_idsnot present arrayoptional - Added
config › limits › protected_ids › [items]not present string - Added
config › schema_versionnot present integer, const=1, default=1optional - Added
config › strategynot present object, additionalProperties=falseoptional - Added
config › strategy › draftnot present string, default="balanced_value", enum (5)optional - Added
config › strategy › seasonnot present string, default="projected_points", enum (3)optional - Added
config › strategy › waivernot present string, default="immediate_starter", enum (3)optional - Added
expected_revisionintegerrequired integerrequired - Description changed
1 unchanged entry
(schema root)
Output schema
Output schema unchanged.
1 unchanged entry
(schema root)
Link to this toolExact 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.
No tools match these filters.
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
krmisystems-fantasy-football-manager/0.3.2~0.3.3The published comparison identifier.- Analyzer
modelara v0.1.8Policy version not recorded.Analyzed Nov 17, 2026, 12:49 AM UTC.- Distribution
pypi- Previous release
fantasy-football-manager@0.3.217 tools · published by the registry Sep 8, 2026, 9:06 PM UTC · upstream 0.3.2Artifact integritysha256-2hhzSmry+60NHNsHOZIS/ckSpMBnZlIF6edze3lY27M=Artifactfantasy_football_manager-0.3.2-py3-none-any.whlInterface fingerprintsha256:366608210502f68c3f92464fab5926fd25e34455c3ea3926bd978cb026e7cce4Exact interface snapshot (JSON)- Current release
fantasy-football-manager@0.3.317 tools · published by the registry Sep 9, 2026, 5:30 AM UTC · upstream 0.3.3Artifact integritysha256-LBWTzHiMsxewAI3djfhB7mhbf1bRkvZ0RXBTAzKbA10=Artifactfantasy_football_manager-0.3.3-py3-none-any.whlInterface fingerprintsha256:1c33d61686117cc9526c608cc659fb91f53e65d7b81214cf17972dad9af128d5Exact interface snapshot (JSON)- Comparison record
- Exact comparison record (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
Policy failures
8 findingsexecute_demo_actionFAILChangedget_power_rankingsFAILChangedimport_league_snapshotFAILChangedload_demoFAILChangedrank_waiver_candidatesFAILChangedrecommend_draftFAILChangedrecommend_lineupFAILChangedupdate_manager_configFAILChanged
Requires review
9 findingsexecute_demo_actionREVIEWChangedShow details
input property confirmation changed: +description="Set true only afte…input property proposal_id changed: +description="Exact proposal_id …
get_action_historyREVIEWChangedShow details
input property limit changed: +description="Maximum audit reco…
import_league_snapshotREVIEWChangedShow details
input property expected_revision changed: +description="Current snapshot r…input property snapshot changed: additionalProperties true → false, +description="Complete league ob…, +properties={"budget":{"anyOf":…, +required=["league_id","team_…, title "Snapshot" → "LeagueSnapshot"
load_demoREVIEWChangedShow details
input property mode changed: +description="Fictional scenario…, +enum=["season","draft"]
prepare_actionREVIEWChangedShow details
input property action changed: +description="Synthetic action: …, +enum=["draft_pick","set_…input property payload changed: -additionalProperties, +anyOf=[{"additionalProper…, +description="Exact action data.…, +examples=[{"player_id":"demo…, -type
rank_waiver_candidatesREVIEWChangedShow details
input property limit changed: +description="Maximum number of …
recommend_draftREVIEWChangedShow details
input property seed changed: +description="Random seed for th…input property trials changed: +description="Simulation trials …
start_draft_monitorREVIEWChangedShow details
input property interval_seconds changed: +description="Pause in seconds a…input property trials changed: +description="Trials per simulat…
update_manager_configREVIEWChangedShow details
input property config changed: additionalProperties true → false, +description="Replacement config…, +properties={"automation":{"add…, title "Config" → "ManagerConfig"input property expected_revision changed: +description="Current config_rev…
Informational
23 findingsexecute_demo_actionINFOChangedget_action_historyINFOChangedget_capabilitiesINFOChangedget_draft_recommendationsINFOChangedget_manager_configINFOChangedget_power_rankingsINFOChangedget_source_statusINFOChangedget_teamINFOChangedimport_league_snapshotINFOChangedload_demoINFOChangedprepare_actionINFOChangedrank_waiver_candidatesINFOChangedrecommend_draftINFOChangedrecommend_lineupINFOChangedstart_draft_monitorINFOChangedstop_draft_monitorINFOChangedupdate_manager_configINFOChangedexecute_demo_actionINFOChangedShow details
declared idempotent: false → true
get_power_rankingsINFOChangedShow details
declared idempotent: true → false
rank_waiver_candidatesINFOChangedShow details
declared idempotent: true → false
recommend_draftINFOChangedShow details
declared idempotent: true → false
recommend_lineupINFOChangedShow details
declared idempotent: true → false
stop_draft_monitorINFOChangedShow details
declared idempotent: false → true
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.