AgentPlaybooks

List playbooks owned by or shared with the authenticated user, including access role and content counts.

Create a new playbook. A playbook is a container for personas (AI personalities), skills (capabilities), and memory (persistent storage).

Get a playbook with its singleton persona, skills, connected MCP servers, and memory.

Delete a playbook and all its contents (personas, skills, memory, API keys). This action cannot be undone!

Set the singleton persona (AI identity and system prompt) for a playbook. Backward-compatible alias for updating persona fields.

Update a persona's name, system prompt, or metadata.

Reset the singleton persona to the default assistant. The playbook always retains one logical persona.

List all skills (capabilities/rules) in this playbook Target a playbook with playbook_id.

Get detailed information about a specific skill Target a playbook with playbook_id.

Read a specific memory entry by key. Automatically increments access count. Memory supports 3 tiers: 'working' (active scratch pad), 'contextual' (recent context), 'longterm' (archived). Use 'hierarch…

Search memories by text, tags, tier, or type. Returns summaries for large memories. Use tags for categorical search; use tier to focus on active vs archived data; use memory_type to find task graphs. …

Write a memory entry. Use tier='working' for active tasks, 'contextual' for background context, 'longterm' for completed work. Set memory_type='hierarchical' and parent_key to build task graphs. Use s…

Delete a memory entry (requires API key) Target a playbook with playbook_id.

Consolidate multiple related memories into a parent memory with summary. Reduces context size while preserving detail access via children. Target a playbook with playbook_id.

Promote a memory to a higher tier or boost its priority for active use. Target a playbook with playbook_id.

Get a context-optimized view of memories. Returns full working memory, summaries for contextual, and keys only for longterm. Target a playbook with playbook_id.

Archive memories from working/contextual to longterm tier. Useful for cleaning up after completing tasks. Target a playbook with playbook_id.

Get hierarchical tree view of memories showing parent-child relationships. Use this to visualize task graphs and track parallel operations. Includes status for each node. Target a playbook with playbo…

Create a hierarchical task plan in one call. Creates a parent 'plan' memory with children for each subtask. Use this for complex multi-threaded work that agent swarms can coordinate on. Each subtask g…

Update the status of a task node in a hierarchical plan. When all children of a parent are 'completed', the parent is auto-updated. Returns the current subtree state. Target a playbook with playbook_i…

List canvas documents in a workflow run. Canvas documents are collaborative markdown files that multiple agents can edit in parallel. Target a playbook with playbook_id.

Read a canvas document. Returns full content, sections structure, and metadata. Optionally read a specific section by ID. Target a playbook with playbook_id.

Create or fully replace a canvas document. Markdown headings are auto-parsed into sections for parallel editing. Use patch_canvas_section for partial updates. Target a playbook with playbook_id.

Edit a specific section of a canvas document. Parallel-safe: only updates the targeted section. Lock the section first for safety in multi-agent scenarios. Target a playbook with playbook_id.

Get the table of contents for a canvas document. Returns section IDs, headings, and levels for navigation and patch_canvas_section. Target a playbook with playbook_id.

Lock a section for exclusive editing. Prevents other agents from modifying it. Remember to unlock when done. Target a playbook with playbook_id.

Unlock a previously locked section so other agents can edit it. Target a playbook with playbook_id.

Create a new skill for this playbook. Use this to expand capabilities. Requires full or skills:write permission. Target a playbook with playbook_id.

Update an existing skill in this playbook. Requires full or skills:write permission. Target a playbook with playbook_id.

Delete a skill from this playbook. Requires full or skills:write permission. Target a playbook with playbook_id.

List historical versions of a skill for auditing or rollback. Target a playbook with playbook_id.

Rollback a skill to a previous version. Requires full or skills:write permission. Target a playbook with playbook_id.

Update the persona/system prompt or the project instructions of this playbook. Handle with extreme care! Requires full or playbooks:write permission. Target a playbook with playbook_id.

List the MCP and OpenAPI servers connected to this playbook, including transport metadata and discovered capability counts. Target a playbook with playbook_id.

Call a tool on one of this playbook's connected MCP servers. This stable wrapper lets the user control plane apply newly created playbooks without dynamically changing its own tool list. Target a play…

Connect an MCP or OpenAPI server to this playbook. Requires playbooks:write or full permission. Target a playbook with playbook_id.

Update a connected MCP or OpenAPI server. Requires playbooks:write or full permission. Target a playbook with playbook_id.

Disconnect an MCP or OpenAPI server from this playbook. Requires playbooks:write or full permission. Target a playbook with playbook_id.

List workflow runs for this playbook. Runs isolate canvas artifacts and execution context. Target a playbook with playbook_id.

Create a workflow run so this playbook can be applied immediately with isolated context and canvas artifacts. Target a playbook with playbook_id.

Update a workflow run's name, status, or context. Target a playbook with playbook_id.

Delete a workflow run and its isolated canvas artifacts. Target a playbook with playbook_id.

List all secret names and metadata in this playbook. Does NOT return values — secret values are never exposed to agents. Requires secrets:read permission. Target a playbook with playbook_id.

Make an HTTP request with a secret injected as a header. The secret value is NEVER returned to the agent — it is decrypted and used server-side only. Use this to authenticate API calls without exposin…

Store a new encrypted secret. The value is encrypted with AES-256-GCM using a per-user derived key and never stored or returned in plaintext. Requires secrets:write permission. Target a playbook with …

Rotate (update) an existing secret with a new value. The old value is permanently replaced and cannot be recovered. Requires secrets:write permission. Target a playbook with playbook_id.

Permanently delete a secret. Cannot be undone. Requires secrets:write permission. Target a playbook with playbook_id.