agent-recall-mcp

Render the project's narrative spine on demand — phase count, day span, sessions,

Full-text search across all journal entries for a project.

Add insights to the awareness system. Call at end of session.

[RETRIEVE — use freely, any time] Use when the user asks to recall, search, find, or look up previous memory, context, or decisions.

Full-text search across all palace rooms. Results are ranked by room salience.

List available journal entries for a project.

Layer 1: lightweight Q&A capture. Appends to today's log file without loading the full journal.

Read a journal entry. Returns the full file content for agent cold-start. Use date='latest' for the most recent entry.

List all projects tracked by agent-recall on this machine.

Return the full content of the currently active project phase, or null if none is open.

[MID-SESSION WRITE — single fact/decision; saying it is not saving it] Use when the user asks to remember, store, note, or save a specific decision, fact, or insight.

[ENTRY — call FIRST, before acting] Use when the user asks to start, load, continue, resume, or open memory for a project. Set mode='lite' for a ≤500-token briefing (good for fresh conversations where the agent will pull memory on demand via recall()).

Read a room overview or specific topic from the Memory Palace.

Move entries older than N days to cold archive. Keeps a one-line summary per archived entry.

Before starting a task, recall cross-project insights that apply.

[MID-SESSION — safe any time; for alignment, before risky decisions] Use when the user asks to validate understanding, verify alignment, or check if their interpretation matches the human's intent. Also call BEFORE a high-risk action — publish, deploy, delete, credential exposure, external send/message, or any other irreversible write — passing `action_description` (one sentence, what you're about to do). Returns matching corrections/rules/insights plus a `verdict`: `blocked` means an authoritative correction OVERRIDES the plan — read it before proceeding.

Search ALL memory stores at once — palace, journal, and insights.

Write a memory to a palace room. Triggers fan-out: cross-references are updated in connected rooms.

Save a permanent IF-THEN behavior commitment for the project. Unlike insights (semantic facts)

Surface a contradiction between the human's current input and a prior statement/decision. Helps the human clarify their own thinking.

Append content to the current journal entry (creates today's file if absent).

Returns a palace-first cold-start package.

Context cache — store or recall pre-computed analysis results.

Layer 1: structured JSON session state. Faster than markdown for cold-start.

List all project phases (milestones) in order — the project's narrative spine.

Save a memory — the system classifies your content and routes it to the right store

Close the currently active project phase with the three reflection fields:

Read lessons from knowledge files. Used before starting work to learn from past mistakes.

Call BEFORE any non-trivial action (publish, push, deploy, schema change, file delete,

[ON SAVE/EXIT — YOU must call this; nothing auto-saves] Use when the user asks to save, checkpoint, summarize, end, retain, or persist the current session.

Record what the agent understood, its confidence, and any human correction. Measures the Intelligent Distance gap.

Progressive context loading for cold-start.

Generate L3 semantic synthesis from recent journals and palace rooms.

Write a structured lesson to the knowledge palace room.

Health check: find stale memories, orphan rooms (no connections), low-salience entries,

Open a new project phase (milestone) in the project's narrative spine.

Condense old daily journals into weekly summaries.

unfenced fixture