Deerdawn

Call at the start of every session. Returns your active project context — current task, recent decisions, open questions — so you can resume without re-explaining the project.

List all your DeerDawn projects with their current task and project ID.

Create a new DeerDawn project to track a codebase, feature, or initiative. Pass parent_project_id to create it as a subproject under an umbrella (e.g. a workstream of a larger product) — subprojects keep their own task and board.

Set which subproject is active for this codebase, so start_session, update_context, and manage_todos target it by default. The choice is remembered per workspace until you switch again. Keyed to the codebase (workspace_key), which the Claude Code / Cursor MCP sends automatically.

Archive a dead or finished project so it stops cluttering list_projects and never auto-resolves as the active project. Preserves all history. Pass unarchive:true to restore it. Refuses to archive an umbrella that still has live subprojects.

Re-parent an existing project: nest it under a top-level umbrella by passing parent_project_id, or promote it back to top-level by omitting parent_project_id. Use this to organize related projects into one umbrella after the fact. One level deep; the umbrella must be top-level and the project being

Rename a project (umbrella or subproject). Use to tidy inconsistent, unclear, or shouty project names — the change is immediate and flows into every future brief. Junk names (file paths, "Untitled") are rejected.

Get the current context for a specific project — task, status, decisions, goals, and open questions. If you have not called start_session yet this session, call it first. Call list_projects if you don't have a project_id.

Update the context for a project. Call this after a decision is made, a task changes, or a feature ships.

Manage the active project's task board — a scrum-style board where every task is a card that moves across phase columns: Backlog → Blocked → In Progress → Integration → Testing → Done. Actions: add (new cards, default to Backlog or a named phase), move (re-column a card), complete (shorthand for → D

Search across all your DeerDawn projects by keyword or topic. If you have not called start_session yet this session, call it first — it may already answer the question without a search.

Get a full overview of your workspace — all projects, durable knowledge, and capabilities.

Return the durable project map: hot paths, entrypoints, named surfaces, and recent debugging notes. Call this BEFORE searching the codebase — the file may already be known. Omit project_id to use the active project.

Search durable path memory (hot paths, surfaces, debugging notes) before doing a repo-wide search. Use a path-oriented query like "auth modal" or "signup route".

Query typed workspace knowledge — services, capabilities, patterns, environments, and workflows — for architecture-level orientation.

Record durable file/folder hot paths AFTER you discover them, so future sessions find them before searching. Append nodes — do not restate the whole tree.

Create or update typed workspace entities (services, capabilities, patterns) and relationships. Use for durable architecture knowledge that should persist beyond a single project note.

Record a durable debugging finding, optionally tied to file paths and a commit/diff ref, after diagnosing a bug or tracing a non-obvious handoff.

Mark a knowledge or decision entity as resolved so it stops appearing in normal context loads. Resolved entries are still retrievable via get_workspace_entities({ status: "resolved" }). Use when a decision has been implemented, a known issue has been fixed, or a knowledge entry is no longer relevant

Return the most relevant next context to fetch or use for the current task/session, including open questions and pending decisions ranked for relevance and token efficiency.

Clean up task context at handoff/finish time. Marks the finished task as completed, clears open-task command history, and returns unresolved questions that still need answers.

List available shared skills (name + description, no body). Use on web tools (ChatGPT, Claude.ai) that cannot read skill files from disk.

Fetch one shared skill in full (including markdown body) by slug. Call after list_skills when a skill is relevant to the current task.