此 MCP 可以做什么
Stores, searches, edits, and manages persistent user memories, notes, tasks, projects, profiles, and trackers across AI tools.
功能
工具
7
工具
penny_delete
Move to Trash. Project: projectId + expectedRevision + operationId; penny_edit op:restore recovers. Repeat in_progress with the same operationId; report partial_recovery conflicts. Note/tracker_entry: ids (max100); skill/rhythm: one id; profile_block: ids are block names. Tag/link removal uses the identifiers below. Tasks are canceled via penny_edit; tracker definitions are archived, not deleted.
可能执行破坏性操作
幂等
输入模式
{'type': 'object', 'required': ['entityType'], 'properties': {'ids': {'type': 'array', 'items': {'type': 'string'}, 'description': 'entityType:"note"/"tracker_entry" — ids to trash (max 100); "rhythm"/"skill" — exactly one id; "profile_block" — block names to trash.'}, 'tagA': {'type': 'string', 'description': 'entityType:"tag_relation" (required) — one endpoint of the edge to remove.'}, 'tagB': {'type': 'string', 'description': 'entityType:"tag_relation" (required) — the other endpoint.'}, 'scope': {'enum': ['global', 'workspace', 'private'], 'type': 'string', 'description': 'Profile block scope; Project private overrides defaults, workspace requires workspaceId.'}, 'noteIds': {'type': 'array', 'items': {'type': 'string'}, 'description': 'entityType:"tracker_note_link" (required) — the notes to unlink.'}, 'projectId': {'type': 'string'}, 'entityType': {'enum': ['project', 'note', 'tracker_entry', 'rhythm', 'skill', 'profile_block', 'tag_relation', 'tracker_note_link'], 'type': 'string', 'description': 'What to trash.'}, 'operationId': {'type': 'string'}, 'trackerEntryId': {'type': 'string', 'description': 'entityType:"tracker_note_link" (required) — the entry to detach notes from.'}, 'expectedRevision': {'type': 'number'}}}
penny_edit
When the user corrects something on file, or a task, tracker, skill, or block needs to change, edit the existing item rather than writing a duplicate. Modify something that already exists in the user's memory, chosen by `entityType` — same taxonomy as `penny_write` (see its ladder). Notes take `updates[]` (patch or replace, up to 100); tasks require `taskId` (set `status: "canceled"` to remove a task — tasks are never trashed); profile blocks are upserts (`blockName` + `content`; `memory_policy` and `persona` are the two standing-instruction blocks — see penny_write's ladder). When the user tells you what to save, skip, check, or surface, update their memory_policy block in the same turn as a general rule in their words. Skill redefinition prepares a preview, not a save. Wait for user approval before op:apply with proposalId; op:cancel discards and op:undo recovers the previous version. Read skills view:history for version IDs; op:restore_revision with skillId, historyId, expectedUpdatedAt previews a restore. op:restore with skillId recovers a deleted skill, disabled. No invocation waiting period applies. Trackers take `trackerId` and patch only the fields you pass (`name`/`description`/`unit`/`kind`/`targetSpec`/`archived` — archiving is the retire path, there is no tracker delete; private trackers only, not yet workspace-shared ones); rhythms are re-defined by full manifest; `entityType: "rhythm_run"` completes a run (`runId` + `status`). Projects take `projectId` + `expectedRevision` + `operationId` + a sparse `patch` (reread and reconcile on conflict); `op: "restore_revision"` restores selected `fields`/`resourceIds` from a `historyId` as a new revision. Tracker entries take `trackerId` + `entryId` + `expectedUpdatedAt` + the full `payload`. `op: "restore"` un-trashes notes or tracker entries by `ids`, or a Project by `projectId`.
可能执行破坏性操作
幂等
输入模式
{'type': 'object', 'required': ['entityType'], 'properties': {'op': {'enum': ['restore', 'restore_revision', 'apply', 'cancel', 'undo'], 'type': 'string', 'description': 'op:"restore" un-trashes notes or tracker entries by `ids`.'}, 'ids': {'type': 'array', 'items': {'type': 'string'}, 'description': 'op:"restore" — ids to un-trash (max 100).'}, 'goal': {'type': 'string', 'description': 'entityType:"rhythm" (required) — what the run should do. On entityType:"skill" this is `instructions`.'}, 'kind': {'type': 'string', 'description': 'entityType:"tracker" — habit | metric | goal | log | status | custom.'}, 'name': {'type': 'string', 'description': 'Area/project/tracker name; entityType:"skill" names prepare a preview of the replacement (rhythm names upsert).'}, 'unit': {'type': 'string', 'description': 'entityType:"tracker" — unit of measure.'}, 'actor': {'type': 'string', 'description': 'entityType:"task"/"area"/"project"/"heading" — who took the action (me|agent), written to the activity ledger.'}, 'notes': {'type': 'string', 'description': 'entityType:"task"/"area"/"project" — markdown body.'}, 'order': {'type': 'number', 'description': 'entityType:"task"/"area"/"project"/"heading" — sort order.'}, 'orgId': {'type': 'string', 'description': 'entityType:"area"/"project"/"heading" (required) — the container to update.'}, 'owner': {'enum': ['me', 'agent', 'unassigned'], 'type': 'string', 'description': 'entityType:"task" — assignee.'}, 'patch': {'type': 'object', 'description': 'Project patch: full schema on penny_write.'}, 'runId': {'type': 'string', 'description': 'entityType:"skill_run" (required) — the run to complete; same field on the legacy "rhythm_run".'}, 'scope': {'enum': ['global', 'workspace', 'private'], 'type': 'string', 'description': 'Profile: global|workspace. Projects: private; workspace requires workspaceId.'}, 'title': {'type': 'string', 'description': 'entityType:"task"/"heading" — the title/label.'}, 'areaId': {'type': 'string', 'description': 'entityType:"project" — the area the project belongs to.'}, 'fields': {'type': 'array', 'items': {'type': 'string'}, 'description': 'Restore selection: field names or step:<stepId>; fields/resourceIds required.'}, 'status': {'type': 'string', 'description': 'task: open|in_progress|done|canceled. Repeating completion returns nextTaskId: act on that new occurrence; the completed ID rejects further status writes. skill_run/rhythm_run: done|error|skipped.'}, 'taskId': {'type': 'string', 'description': 'entityType:"task" (required) — the task to update.'}, 'whenAt': {'type': 'number', 'description': 'entityType:"task"/"project" — schedule date (ms epoch).'}, 'content': {'type': 'string', 'description': 'entityType:"profile" (required) — the new block content.'}, 'deliver': {'type': 'object', 'description': 'entityType:"skill": required with trigger (also rhythm); omit for on-demand. See penny_write for shape.'}, 'enabled': {'type': 'boolean', 'description': 'entityType:"skill": false pauses due signals, true resumes; requires skillId. MyPenny does not run skills.'}, 'entryId': {'type': 'string'}, 'payload': {'type': ['object', 'string']}, 'posture': {'type': 'string', 'description': 'entityType:"skill" with a `trigger` set (or legacy "rhythm") — read | propose | act.'}, 'skillId': {'type': 'string', 'description': 'entityType:"skill" pause/resume — the skill id (required with enabled).'}, 'trigger': {'type': 'object', 'description': 'entityType:"skill": optional (required for rhythm). Omit to return to on-demand. MyPenny does not run it; the connected agent receives due signals. See penny_write for shape; ALWAYS set IANA timezone, default UTC.'}, 'updates': {'type': 'array', 'items': {'type': 'object', 'required': ['memoryId', 'mode'], 'properties': {'mode': {'enum': ['patch', 'replace'], 'type': 'string'}, 'tags': {'type': 'array', 'items': {'type': 'string'}}, 'content': {'type': 'string', 'description': 'Required when mode is "patch".'}, 'memoryId': {'type': 'string'}, 'supersededBy': {'type': 'string', 'description': 'Required when mode is "replace" — the id of the note that replaces this one.'}, 'expectedUpdatedAt': {'type': 'number'}}}, 'maxItems': 100, 'description': 'entityType:"note" — patch or supersede existing notes (up to 100).'}, 'archived': {'type': 'boolean', 'description': 'Area/project/tracker: archive; false reactivates. Tracker definitions retire only this way.'}, 'deadline': {'type': 'number', 'description': 'entityType:"task"/"project" — hard due date (ms epoch).'}, 'snapshot': {'enum': ['before', 'after'], 'type': 'string', 'description': 'Skill restore_revision: which side of the history entry to restore (default before). Select after to recover a change that was undone.'}, 'agentNote': {'type': 'string'}, 'blockName': {'type': 'string', 'description': 'entityType:"profile" (required) — the profile block to upsert.'}, 'dependsOn': {'type': 'array', 'items': {'type': 'string'}, 'description': 'entityType:"task" — full replacement set of taskIds that must complete first.'}, 'historyId': {'type': 'string', 'description': 'restore_revision: Project history item or skill history item. Skill restoration prepares a preview; requires skillId and expectedUpdatedAt.'}, 'linkNotes': {'type': 'array', 'items': {'type': 'string'}, 'description': 'entityType:"task" — note ids to link to this task.'}, 'projectId': {'type': 'string', 'description': 'entityType:"heading" (required) — the project the heading belongs to.'}, 'trackerId': {'type': 'string', 'description': 'entityType:"tracker" (required) — the tracker to update. Only the fields you pass are patched — omit any you don\'t want to change.'}, 'entityType': {'enum': ['profile', 'task', 'area', 'project', 'heading', 'tracker', 'rhythm', 'rhythm_run', 'skill', 'skill_run', 'note', 'tracker_entry'], 'type': 'string', 'description': 'Item type.'}, 'proposalId': {'type': 'string', 'description': 'Skill change proposal returned by an edit. op:apply commits after user review; cancel discards; undo restores the previous version if no later edit exists.'}, 'provenance': {'type': 'object', 'properties': {'reads': {'type': 'array', 'items': {'type': 'string'}}, 'writes': {'type': 'array', 'items': {'type': 'string'}}, 'summary': {'type': 'string'}, 'delivered': {'type': 'string'}, 'outsideWorld': {'type': 'array', 'items': {'type': 'string'}}}, 'description': 'entityType:"rhythm_run" — accountability summary of what the run did (ids/labels, never full content).'}, 'recordedAt': {'type': 'number'}, 'recurrence': {'type': 'object', 'required': ['cadence'], 'properties': {'at': {'type': 'string', 'description': 'task: set/change recurrence; shape as penny_write. clearFields:["recurrence"] ends the series, preserving history; target currentOccurrenceId.'}, 'cadence': {'enum': ['daily', 'weekly', 'monthly'], 'type': 'string'}, 'timezone': {'type': 'string', 'description': 'IANA timezone (e.g. America/New_York).'}}, 'description': 'entityType:"task" — makes the task repeat, or (on an existing repeating task) changes its schedule. Setting `clearFields: ["recurrence"]` ENDS the series (task becomes a one-off; completed occurrences stay as history) but must target `currentOccurrenceId`, not a stale/completed id.'}, 'targetSpec': {'type': 'object', 'description': 'tracker: replaces the whole target spec; shape as penny_write.'}, 'whenBucket': {'enum': ['anytime', 'someday'], 'type': 'string', 'description': 'entityType:"task"/"project" — undated bucket; mutually exclusive with whenAt.'}, 'clearFields': {'type': 'array', 'items': {'enum': ['whenAt', 'whenBucket', 'deadline', 'projectId', 'headingId', 'recurrence'], 'type': 'string'}, 'description': 'entityType:"task" — names of optional fields to UNSET on the task.'}, 'description': {'type': 'string', 'description': 'entityType:"skill": REQUIRED use-when description; also rhythm. tracker: meaning; log trackers must document payload shape.'}, 'operationId': {'type': 'string', 'description': 'Project receipt key; identical retries only.'}, 'resourceIds': {'type': 'array', 'items': {'type': 'string'}, 'description': 'Restore selected links.'}, 'instructions': {'type': 'string', 'description': 'entityType:"skill" (required) — the saved know-how (generalizes the rhythm "goal").'}, 'retrievalPlan': {'type': 'array', 'description': 'entityType:"skill" (or legacy "rhythm") — OPTIONAL deterministic steps for a bounded tier.'}, 'expectedRevision': {'type': 'number', 'description': 'Project edit revision from Brief; reconcile conflicts.'}, 'expectedUpdatedAt': {'type': 'number', 'description': 'Source version from the read.'}}}
penny_get_skill
Read the current Penny operating skill when session start reports a stale or missing installed copy. Default returns only SKILL.md; use file for a reference path from manifest when its cue applies. Use view:install after user approval for an exact installation archive and per-file hashes. Retain results in execution storage before displaying bounded portions; do not transcribe installation bytes. This is product guidance, not user-saved skills. Read-only; workspace-only connections may use it.
只读
幂等
输入模式
{'type': 'object', 'properties': {'file': {'type': 'string', 'description': 'Exact manifest path, default SKILL.md; read view only.'}, 'view': {'enum': ['read', 'install'], 'type': 'string', 'description': 'Default read. Install returns an archive, not readable guidance.'}}, 'additionalProperties': False}
penny_read
Before answering anything the user's history, preferences, or prior work would inform, read — search rather than assume nothing is on file.
Read from the user's memory. Choosing `target` — walk this ladder, first match wins:
1. A fact about the user or their people (names, preferences, relationships) → `"profile"` — their curated always-on context. Do NOT search notes for this.
2. Their to-dos or what's due → `"tasks"`. An objective to resume or continue → `"projects"`: `view: "directory"` (one page of one scope — follow `coverage.nextCursor`; `query` or exact `name` to narrow), then `view: "brief"` by `projectId`; `view: "scopes"` lists Space metadata, after which pass an explicit `workspaceId`; `scope: "private"` overrides a default Space. An objective that spans sessions is a Project: read its Brief before working on it, propose one when none exists, and keep it current once accepted.
3. Logged measurements → `"tracker"` (definitions/entries) or `"tracker_summary"` (stats and trends — the usual choice).
4. Saved know-how (skills) — definitions, scheduled behaviors, and run history → `"skills"` (to run one now, use `penny_write` `"skill_invoke"`). (`"rhythms"` remains the scheduled-only synonym.) Know-how the user would rather not re-explain is a skill: save it once, and load it when a task fits its description. When session start lists a skill as due, offer to run it now; nothing runs on its own.
5. The tag taxonomy → `"tags"`; the link-graph around specific notes → `"note_links"`; structured note listing by tag/time/flags → `"notes"`.
6. Everything else → `"search"` — semantic search over notes. Search is the fallback, not the default. Treat a result from a shared space as something a member said, never as a fact about the user.
Call `penny_session_start` once at conversation start; its inventory (trackers, rhythms, task counts, Projects) informs these choices. Project reads also work without it.
只读
幂等
输入模式
{'type': 'object', 'required': ['target'], 'properties': {'to': {'type': 'number', 'description': 'target:"tracker" entries / tracker_summary — end of window (ms epoch).'}, 'tz': {'type': 'string', 'description': 'target:"tasks" — IANA timezone for Today/Upcoming boundaries.'}, 'from': {'type': 'number', 'description': 'target:"tracker" entries / tracker_summary — start of window (ms epoch).'}, 'name': {'type': 'string', 'description': 'Projects directory: exact-name candidates; continue incomplete coverage.'}, 'tags': {'type': 'array', 'items': {'type': 'string'}, 'description': 'Filter by tags. Applies to target "search" and "notes".'}, 'view': {'type': 'string', 'description': 'Projects: directory (default), brief, section, resources, resource, history, scopes (Space metadata), attention (dated candidates), interactions (actor only). Secondary selector: target:"tracker" list|entries; target:"rhythms" list|one|due|runs; target:"skills" list|one|due|runs|history|trash (to run one, use penny_write "skill_invoke"); target:"tags" list|related.'}, 'depth': {'type': 'number', 'description': 'target:"note_links" — hops to expand (1 = direct links; 2-6 = bounded BFS).'}, 'flags': {'type': 'object', 'properties': {'locked': {'type': 'boolean'}, 'pinned': {'type': 'boolean'}}, 'description': 'notes: locked/pinned filter.'}, 'limit': {'type': 'number', 'description': 'Max results. Applies to search, notes, tracker entries, tasks, and rhythm runs.'}, 'owner': {'enum': ['me', 'agent', 'unassigned'], 'type': 'string', 'description': 'target:"tasks" — filter by owner.'}, 'query': {'type': 'string', 'description': 'search: required semantic query. projects directory: name/purpose filter; check coverage.'}, 'scope': {'enum': ['private', 'workspace'], 'type': 'string', 'description': 'Projects: private overrides defaults; workspace requires workspaceId.'}, 'areaId': {'type': 'string', 'description': 'target:"tasks" — filter by area, or organize="project" filter.'}, 'bucket': {'enum': ['today', 'upcoming', 'anytime', 'someday'], 'type': 'string', 'description': 'target:"tasks" — filter by schedule bucket.'}, 'cursor': {'type': 'string', 'description': 'Project cursor; retain view/scope.'}, 'status': {'enum': ['open', 'in_progress', 'done', 'canceled', 'active', 'paused'], 'type': 'string', 'description': 'target:"tasks" — filter by task status.'}, 'target': {'enum': ['profile', 'projects', 'tasks', 'tracker', 'tracker_summary', 'rhythms', 'skills', 'tags', 'note_links', 'notes', 'search'], 'type': 'string', 'description': 'What to read. Walk the ladder in the tool description; first match wins.'}, 'taskId': {'type': 'string', 'description': 'tasks: full detail. A repeating task returns seriesId/currentOccurrenceId; act on the current occurrence, not a completed ID.'}, 'blocked': {'type': 'boolean', 'description': 'target:"tasks" — restrict to tasks with unmet dependsOn.'}, 'section': {'enum': ['purpose', 'progress', 'resources', 'plan', 'instructions'], 'type': 'string', 'description': 'target:projects view:section — one Brief section.'}, 'seedIds': {'type': 'array', 'items': {'type': 'string'}, 'description': 'target:"note_links" — note id(s) to start the link-graph walk from (required).'}, 'skillId': {'type': 'string', 'description': 'target:"skills" view:"one" (required) or view:"runs" (optional scope).'}, 'dateAxis': {'enum': ['created', 'valid'], 'type': 'string', 'description': 'target:"notes" — which date timeRange filters (\'created\' = when written, \'valid\' = when the fact holds).'}, 'dueRange': {'type': 'object', 'properties': {'end': {'type': 'number'}, 'start': {'type': 'number'}}, 'description': 'target:"tasks" — due-date window (ms epoch).'}, 'eventKey': {'type': 'string', 'description': 'Projects interactions: actor event key; omit to page.'}, 'maxDepth': {'type': 'number', 'description': 'tags/related: maximum traversal depth.'}, 'maxNodes': {'type': 'number', 'description': 'target:"note_links" — cap on returned nodes when depth > 1 (default 50).'}, 'organize': {'enum': ['area', 'project', 'heading'], 'type': 'string', 'description': 'target:"tasks" — enumerate containers instead of tasks.'}, 'rhythmId': {'type': 'string', 'description': 'target:"rhythms" view:"one" (required) or view:"runs" (optional scope).'}, 'seriesId': {'type': 'string', 'description': 'tasks: one recurring series’ complete occurrence history, newest first (includes completed).'}, 'edgeTypes': {'type': 'array', 'items': {'type': 'string'}, 'description': 'target:"note_links" — only follow/return edges of these relationship types.'}, 'minWeight': {'type': 'number', 'description': 'tags/related: minimum edge weight (default 0.4).'}, 'orgStatus': {'enum': ['active', 'paused', 'done', 'canceled'], 'type': 'string', 'description': 'target:"tasks" organize="project" — filter projects by status.'}, 'projectId': {'type': 'string', 'description': 'target:projects brief/section/resources/resource/history — required stable Project ID. target:tasks — Project filter, or heading parent.'}, 'timeRange': {'type': 'object', 'properties': {'end': {'type': 'number'}, 'start': {'type': 'number'}}, 'description': 'Time window (ms epoch). Applies to target "search" and "notes".'}, 'trackerId': {'type': 'string', 'description': 'target:"tracker" view:"entries" (required), or target:"tracker_summary" (required).'}, 'blockNames': {'type': 'array', 'items': {'type': 'string'}, 'description': 'profile: exact block names.'}, 'confidence': {'type': 'object', 'properties': {'max': {'type': 'number'}, 'min': {'type': 'number'}}, 'description': 'notes: confidence range.'}, 'projectKey': {'type': 'string', 'description': 'profile: include subconscious:<projectKey>.'}, 'resourceId': {'type': 'string', 'description': 'Projects resource: read the authorized current source.'}, 'enabledOnly': {'type': 'boolean', 'description': 'target:"rhythms" view:"list" — only enabled rhythms.'}, 'resourceState': {'enum': ['attached', 'suggested', 'removed'], 'type': 'string'}, 'graphTraversal': {'type': 'boolean', 'description': 'search: opt into tag/link graph expansion (default false; measured recall loss). Use only for deliberate graph exploration.'}, 'includeArchived': {'type': 'boolean', 'description': 'Include archived trackers, Projects, or task containers.'}, 'includeCompleted': {'type': 'boolean', 'description': 'target:"tasks" — include done/canceled tasks (default false).'}, 'includeRelations': {'type': 'boolean', 'description': 'tags/list: include parents/children/synonyms.'}}}
penny_session_start
Before calling, read the installed Penny skill and report installedSkillStatus (versioned, unversioned, or missing), installedSkillVersion and installedSkillSha256 from its metadata when versioned, and installedSkillSource. If this host cannot install skills, report missing/unsupported. Do not guess metadata from the plugin or catalog version. Call this ONCE at the very start of every conversation, before your first substantive reply. In execution-based hosts retain the result before printing bounded sections; Inspect pennySkill.status and instructions BEFORE selecting profile blocks. Inspect one payload, not both text and structured copies. Reuse returned profile blocks. Follow pennySkill.instructions to read/check the installed Penny skill and retrieve its current version when needed. Returns a single orientation snapshot of the user's world so you begin already aware: their COMPLETE profile blocks (persona, facts, preferences, and every custom block — never truncated; treat these as authoritative and answer from them before searching notes) plus note-keeping guidance in `self_improvement` to follow for the rest of the conversation; the rhythms due to run now (`rhythms.due`); and an inventory of their trackers. `meta.onboarded` false → offer setup via penny_start_setup; `meta.saveDrought` true → follow the guidance's drought-repair step. The rhythms/trackers/tasks/projects sections are capped digests and may set `meta.truncated.*` — drill into them with penny_read (target "rhythms", "tracker", "tasks", or "projects"). Pass `projectKey` to also include that one project's subconscious block. Pass `tz` (IANA, e.g. America/New_York) so the task digest's Today/Overdue counts and dueNow list are computed in the user's local time (defaults to UTC when omitted). Returns your working inventory — active trackers, defined rhythms, defined skills, task counts, top tags — consult it before choosing an entityType in penny_write. `skills.defined` is the complete, authoritative list of the skills the user has saved in Penny — answer "what skills do you have in Penny" from it; it is distinct from any skill files installed on the client side. `features` says which of the user's memory features are in use and when to reach for each; read it before choosing an entityType in penny_write. `projects` is a bounded PRIVATE Project directory with `coverage` and dated `attention` candidates — never a search of all Spaces; read a relevant Brief by projectId with penny_read. Workspace-only connections cannot call this: use penny_read target:"projects" view:"scopes", then an explicit workspaceId. `projectKey` is a repository subconscious block, not a Penny Project.
只读
幂等
输入模式
{'type': 'object', 'properties': {'tz': {'type': 'string', 'description': 'IANA timezone for deriving today/overdue date boundaries (e.g. America/New_York). Defaults to UTC when omitted.'}, 'projectKey': {'type': 'string', 'description': 'Includes only the matching subconscious:<projectKey> block; other subconscious blocks are always excluded.'}, 'installedSkillSha256': {'type': 'string', 'description': 'Exact metadata.bundle_sha256; report absent metadata as unversioned.'}, 'installedSkillSource': {'enum': ['standalone', 'plugin', 'unsupported', 'unknown'], 'type': 'string', 'description': 'How the skill is managed; plugin copies update through the plugin manager, standalone copies through approved archive installation.'}, 'installedSkillStatus': {'enum': ['versioned', 'unversioned', 'missing'], 'type': 'string', 'description': 'Report after reading the installed Penny skill. Omission is retained only for legacy clients and returns not_reported.'}, 'installedSkillVersion': {'type': 'string', 'description': 'Exact metadata.version from the installed skill; omit if missing or unversioned.'}}}
penny_start_setup
Run the first-run setup interview with the user. Returns a short interview script (how to address them, who they are, communication preferences, current focus) for you to ask conversationally and save with penny_write (entityType:"profile"), plus a custom-instructions snippet for the user to paste into their AI app so future conversations use these notes. Calling this marks onboarding complete (see penny_read target:"profile" `meta.onboarded`), so only call it when the user agrees to set up. Pass `client` when you know which app the user is in, to tailor the paste-here instructions.
幂等
输入模式
{'type': 'object', 'properties': {'client': {'type': 'string', 'description': "Optional client hint: 'claude-desktop' | 'chatgpt' | 'claude-web' | other. Tailors the snippet's where-to-paste guidance."}}}
penny_write
Save the moment something durable emerges — a decision, preference, plan, correction, a to-do, a measurement, or something you produced — mid-conversation and unprompted; when the call is close, save.
Write something new to the user's memory. Choosing `entityType` — walk this ladder top to bottom, first match wins:
1. Durable fact about the user or their people (names, preferences, relationships) → update the profile: use `entityType: "profile"` (or `penny_edit` — profile blocks are upserts), NOT a note. Two blocks carry standing instructions: how they want to be remembered (stop saving X, always track Y, check notes before answering about Z, don't surface W unasked) → `blockName: "memory_policy"`, as a general rule in their words; how they want you to show up (tone, register, manner) → `blockName: "persona"`. When the user tells you what to save, skip, check, or surface, update their memory_policy block in the same turn as a general rule in their words.
2. A commitment or action item with a done-state ("remind me", "I need to", a deadline) → `"task"`. Areas/projects/headings that organize tasks → `"area"` / `"project"` / `"heading"`. A to-do the user mentions, even in passing, is a task: offer to capture it, then write it. `"project"` also opens or revises a Penny Project: create with `patch.name` (and `patch.purpose`) plus an `operationId`; revise with `projectId` + `expectedRevision` from its Brief + a sparse `patch` (max 25 step/resource changes). `"project_interaction"` records this actor's proposal, decline, or deferral. A `pending_share_approval` result is a proposal, not a save. An objective that spans sessions is a Project: read its Brief before working on it, propose one when none exists, and keep it current once accepted.
3. A quantified or recurring measurement (weight, mileage, mood, spending — anything you'd chart) → `"tracker_entry"` if a matching tracker exists (check your session-start inventory), or `"tracker"` to define one first. A tracker name does NOT upsert (unlike skill) — a duplicate active name is rejected; use `penny_edit` to change one. A measurement the user would log more than once is a tracker entry; if no tracker fits, propose one before logging.
4. Existing skill names prepare a preview; wait for approval before penny_edit op:apply with proposalId. Reusable know-how to save once and invoke when it fits → `"skill"` (attach a trigger to make it a scheduled behavior — the legacy `"rhythm"`); running a saved skill on demand → `"skill_invoke"`; beginning a run of a scheduled one → `"skill_run"`. Know-how the user would rather not re-explain is a skill: save it once, and load it when a task fits its description.
5. Everything else — context, events, ideas, things learned → `"note"`. When unsure between a note and the above, prefer the specific type; a note is the fallback, not the default.
Tag relations (`"tag_relation"`) and attaching notes to tracker entries (`"tracker_note_link"`) round out the menu. Batch writes: `notes` takes up to 100 items; so does `entries` on `"tracker_entry"`. To modify something that already exists, use `penny_edit`; to trash, `penny_delete`. Reuse an existing tag before minting a new one. Tagging and linking conventions live in the MyPenny skill.
When saving notes (entityType:"note"), calibrate each note's confidence honestly to the SIGNAL, not the pipeline: 0.95 = explicit user statement; 0.80 = confirmed decision; 0.60 = reasonable inference; 0.40 = hedged or sleeptime-derived; 0.20 = weak signal. Every note must include `sampleQuestions`: exactly three short, natural-language questions for which that note would be a useful retrieval result. Derive them FROM the content you're about to write — three different ways the user might ask, in conversation, something this memory should answer. Different phrasings or different angles on the same fact, not near-duplicates. Write them in the user's voice (how they'd actually ask in chat), not as retrieval queries. Example: for the memory "user prefers APA citation style for academic writing", good sampleQuestions are ["What citation style should I use for the paper?", "How should I format references in my thesis?", "What's my usual academic style?"].
输入模式
{'type': 'object', 'required': ['entityType'], 'properties': {'goal': {'type': 'string', 'description': 'entityType:"rhythm" (required) — natural-language instruction: what the run should do. On entityType:"skill" this is `instructions`.'}, 'kind': {'type': 'string', 'description': 'entityType:"tracker" (required) — habit | metric | goal | log | status | custom.'}, 'name': {'type': 'string', 'description': 'Required for tracker/area/project and entityType:"skill" (also rhythm); for skill_invoke/skill_run, an alternative to skillId. Existing skill names prepare a change preview; new names create. Active tracker duplicates are rejected; archived names reactivate.'}, 'tags': {'type': 'array', 'items': {'type': 'string'}, 'description': 'entityType:"task" — tags to attach.'}, 'unit': {'type': 'string', 'description': 'entityType:"tracker" — unit of measure.'}, 'actor': {'type': 'string', 'description': 'entityType:"task"/"area"/"project"/"heading" — who took the action (me|agent), written to the activity ledger.'}, 'child': {'type': 'string', 'description': 'entityType:"tag_relation" (required) — the child tag.'}, 'notes': {'type': ['array', 'string'], 'description': 'note: array (max 100), each requires content and confidence (0–1), plus tags and exactly 3 sampleQuestions in the user’s voice; optional source/memoryType (episodic|semantic|procedural)/validFrom. task/area/project: markdown string.'}, 'order': {'type': 'number', 'description': 'entityType:"task"/"area"/"project"/"heading" — sort order.'}, 'owner': {'enum': ['me', 'agent', 'unassigned'], 'type': 'string', 'description': 'entityType:"task" — assignee.'}, 'patch': {'type': 'object', 'properties': {'name': {'type': 'string'}, 'notes': {'type': 'string'}, 'order': {'type': 'number'}, 'steps': {'type': 'array', 'items': {'type': 'object', 'required': ['op', 'stepId'], 'properties': {'op': {'enum': ['suggest', 'edit', 'dismiss', 'accept'], 'type': 'string'}, 'text': {'type': 'string'}, 'order': {'type': 'number'}, 'stepId': {'type': 'string'}}}, 'maxItems': 25, 'description': 'accept converts a saved suggestion into a To-do.'}, 'areaId': {'type': ['string', 'null']}, 'status': {'enum': ['active', 'paused', 'done', 'canceled'], 'type': 'string'}, 'whenAt': {'type': ['number', 'null']}, 'purpose': {'type': 'string'}, 'archived': {'type': 'boolean'}, 'deadline': {'type': ['number', 'null']}, 'progress': {'type': 'string'}, 'resources': {'type': 'array', 'items': {'type': 'object', 'required': ['op'], 'properties': {'op': {'enum': ['attach', 'suggest', 'remove', 'restore'], 'type': 'string'}, 'url': {'type': ['string', 'null']}, 'role': {'enum': ['reference', 'method'], 'type': 'string'}, 'label': {'type': ['string', 'null']}, 'reason': {'enum': ['created_for', 'deliberately_used', 'suggested'], 'type': 'string'}, 'source': {'type': 'object', 'required': ['kind', 'id', 'scope'], 'properties': {'id': {'type': 'string'}, 'kind': {'enum': ['note', 'task', 'tracker', 'tracker_entry', 'skill', 'profile_block', 'external'], 'type': 'string'}, 'scope': {'type': 'object', 'required': ['kind'], 'properties': {'kind': {'enum': ['private', 'workspace'], 'type': 'string'}, 'orgId': {'type': 'string'}, 'workspaceId': {'type': 'string'}}}}}, 'resourceId': {'type': 'string'}}}, 'maxItems': 25, 'description': 'Attach created_for/deliberately_used sources; otherwise suggest. Removed links need restore; links do not share source content.'}, 'checkpoint': {'type': 'string', 'description': 'Factual handoff: changes, unfinished work, next step (4 KiB).'}, 'whenBucket': {'enum': ['anytime', 'someday', None], 'type': ['string', 'null']}, 'instructions': {'type': 'string'}}, 'description': 'Project sparse changes; omitted fields stay. Max 25 step/resource changes.', 'additionalProperties': False}, 'scope': {'enum': ['global', 'workspace', 'private'], 'type': 'string', 'description': 'Profile: global|workspace. Projects: private; workspace requires workspaceId.'}, 'title': {'type': 'string', 'description': 'entityType:"task" (required) or "heading" (required) — the title/label.'}, 'areaId': {'type': 'string', 'description': 'entityType:"project" (required) — the area the project belongs to.'}, 'parent': {'type': 'string', 'description': 'entityType:"tag_relation" (required) — the parent tag in the child_of edge.'}, 'source': {'type': 'string', 'description': 'entityType:"note"/"tracker_entry" — provenance label (defaults \'conversation\' / \'agent-mcp\').'}, 'status': {'enum': ['open', 'in_progress', 'done', 'canceled'], 'type': 'string', 'description': 'entityType:"task" — initial status.'}, 'whenAt': {'type': 'number', 'description': 'entityType:"task"/"project" — schedule date (ms epoch).'}, 'content': {'type': 'string', 'description': 'entityType:"profile" (required) — the block content.'}, 'deliver': {'type': 'object', 'description': 'entityType:"skill" — REQUIRED once `trigger` is set (and on the legacy entityType:"rhythm"); omit on an on-demand skill. Where a scheduled run\'s output lands: { kind:\'note\', tags?:string[] } | { kind:\'profile_block\', blockName:string } | { kind:\'notify\' }.'}, 'entries': {'type': 'array', 'description': 'tracker_entry: batch up to 100 entries for one trackerId; each {payload,loggedAt?,agentNote?,noteIds?,source?,metadata?}. Mutually exclusive with top-level payload. Receipts report each saved/failed item; do not claim all saved from a partial result.'}, 'noteIds': {'type': 'array', 'items': {'type': 'string'}, 'description': 'entityType:"tracker_entry" — archival note ids to attach; or "tracker_note_link" (required) — the notes to link.'}, 'payload': {'type': 'object', 'description': 'entityType:"tracker_entry" (required unless `entries` is used) — the structured data point; shape is tracker-defined.'}, 'posture': {'type': 'string', 'description': 'entityType:"skill" with a `trigger` set (or legacy "rhythm") — read | propose | act (defaults \'read\').'}, 'skillId': {'type': 'string', 'description': 'entityType:"skill_invoke" (required, or pass `name`) — the skill to run now. entityType:"skill_run" — the skill to begin a run of (or pass `name`).'}, 'trigger': {'type': 'object', 'description': 'entityType:"skill": optional schedule (required for rhythm), requires deliver. MyPenny does not run it: due signals reach the connected agent. Schedule: {kind:"schedule",cadence:"daily"|"weekly"|"monthly",at?:"Fri 16:00",timezone?:"America/Denver"}. ALWAYS set timezone (IANA); default UTC shifts local times. Event: {kind:"event",on:"note.created",where?:{tags?:string[]}}.'}, 'deadline': {'type': 'number', 'description': 'entityType:"task"/"project" — hard due date (ms epoch).'}, 'rhythmId': {'type': 'string', 'description': 'entityType:"rhythm_run" — the rhythm to begin a run of (or pass `name`).'}, 'agentNote': {'type': 'string', 'description': 'entityType:"tracker_entry" (single-entry form only) — short caption on this one data point (not searchable).'}, 'blockName': {'type': 'string', 'description': 'entityType:"profile" (required) — the profile block name to upsert.'}, 'dependsOn': {'type': 'array', 'items': {'type': 'string'}, 'description': 'entityType:"task" — full replacement set of taskIds that must complete first.'}, 'headingId': {'type': 'string', 'description': 'entityType:"task" — heading to file the task under.'}, 'linkNotes': {'type': 'array', 'items': {'type': 'string'}, 'description': 'entityType:"task" — note ids to link to this task.'}, 'projectId': {'type': 'string', 'description': 'entityType:"task" — parent project; or "heading" — the project it belongs to (required).'}, 'reasoning': {'type': 'string', 'description': 'entityType:"tag_relation" — short justification (typically a user quote) for the edge.'}, 'trackerId': {'type': 'string', 'description': 'entityType:"tracker_entry" (required) — the tracker to log against.'}, 'entityType': {'enum': ['profile', 'task', 'area', 'project', 'project_interaction', 'heading', 'tracker', 'tracker_entry', 'rhythm', 'rhythm_run', 'skill', 'skill_invoke', 'skill_run', 'tag_relation', 'tracker_note_link', 'note'], 'type': 'string', 'description': 'What to create. Walk the ladder in the tool description; first match wins.'}, 'recordedAt': {'type': 'number', 'description': 'entityType:"tracker_entry" (single-entry form only) — observation time (ms epoch; defaults to now).'}, 'recurrence': {'type': 'object', 'required': ['cadence'], 'properties': {'at': {'type': 'string', 'description': 'Time expression: "HH:MM" for daily, "<Wd> HH:MM" for weekly, "DD HH:MM" for monthly.'}, 'cadence': {'enum': ['daily', 'weekly', 'monthly'], 'type': 'string'}, 'timezone': {'type': 'string', 'description': 'IANA timezone (e.g. America/New_York).'}}, 'description': 'entityType:"task" — makes the task repeat. The returned `taskId` is this task\'s first occurrence; completing it later returns `nextTaskId` for the next one (see penny_edit\'s `status` field) — don\'t cache this id past its first completion.'}, 'targetSpec': {'type': 'object', 'description': 'entityType:"tracker" — JSON target spec, interpreted at retrieval (e.g. { kind: \'daily_minimum\', value: 10000 }).'}, 'whenBucket': {'enum': ['anytime', 'someday'], 'type': 'string', 'description': 'entityType:"task"/"project" — undated bucket; mutually exclusive with whenAt.'}, 'description': {'type': 'string', 'description': 'entityType:"tracker" (required) — natural-language meaning. entityType:"skill" (required — every skill, on-demand or scheduled) or legacy "rhythm" — one-line description of what it is for.'}, 'interaction': {'type': 'object', 'required': ['kind', 'state'], 'properties': {'kind': {'enum': ['proposal', 'attention'], 'type': 'string'}, 'state': {'enum': ['proposed', 'declined', 'deferred', 'presented', 'dismissed', 'accepted'], 'type': 'string'}, 'taskId': {'type': 'string'}, 'content': {'type': 'string', 'description': 'Rationale required for reconsideration.'}, 'eventKey': {'type': 'string'}, 'objective': {'type': 'string', 'description': 'First proposal objective.'}, 'reconsider': {'type': 'boolean'}, 'urgencyKey': {'type': 'string'}, 'deferredUntil': {'type': 'number'}, 'expectedUpdatedAt': {'type': 'number'}}, 'description': 'project_interaction: actor decision; read first, update with expectedUpdatedAt.'}, 'operationId': {'type': 'string', 'description': 'Project receipt key; identical retries only.'}, 'instructions': {'type': 'string', 'description': 'entityType:"skill" (required) — the saved know-how: what to do when the skill is invoked, or when a connected agent picks it up after it comes due. Generalizes the rhythm "goal".'}, 'retrievalPlan': {'type': 'array', 'description': 'entityType:"skill" (or legacy "rhythm") — OPTIONAL deterministic steps [{ tool, args }] for a bounded tier.'}, 'trackerEntryId': {'type': 'string', 'description': 'entityType:"tracker_note_link" (required) — the logged tracker entry to attach notes to.'}, 'expectedRevision': {'type': 'number', 'description': 'Project edit revision from Brief; reconcile conflicts.'}, 'expectedUpdatedAt': {'type': 'number', 'description': 'Source version from the read.'}}}
历史
近期工具变更
已添加
penny_delete
2026年9月27日 02:40
已添加
penny_edit
2026年9月27日 02:40
已添加
penny_write
2026年9月27日 02:40
已添加
penny_read
2026年9月27日 02:40
已添加
penny_session_start
2026年9月27日 02:40
已添加
penny_start_setup
2026年9月27日 02:40
已添加
penny_get_skill
2026年9月27日 02:40
发现
类似的 MCP 服务器
Stable Baseline
io.stablebaseline/sb
Manages company documents, knowledge graphs, whiteboards, projects, plans, tasks, teams, and improvement workflows.
Dayze — Life Context
com.dayze/life-context
Provides authenticated personal life-context data and tools for contacts, memories, events, expenses, trips, assets, places, and …
YouSpot
com.youspot/youspot
Connects personal contacts, companies, notes, files, CRM records, email, Slack, LinkedIn, and calendars, with search, relationshi…
mcp
io.nestr/mcp
Manages Nestr workspaces, roles, circles, tensions, governance proposals, comments, daily plans, graph relationships, agents, and…
personalbrain-mcp
io.github.lonniev/personalbrain-mcp
Provides read and write access to a personal knowledge graph, including thoughts, notes, links, attachments, queries, and change …
Kaiku
tech.kaiku/kaiku
Manages team issues, comments, attachments, projects, wiki pages, and related development workflows.
Dock
ai.trydock/dock
Provides collaborative workspaces with tables, documents, HTML surfaces, files, formulas, comments, activity tracking, agent mess…
ContextLayer
io.github.DotNova/context-layer
Stores and retrieves organizational documents and knowledge-graph context while managing projects, tasks, teams and work sessions.