agentleFS
Sign inStart free
arnaugonzalezarnaugonzalez/memory-boost/skills/memory-boost/SKILL.md

memory-boost

Skill0 starsChanged 47 days agoMITView on GitHub

Router for the memory-boost CLI (markdown wiki for agents + drift reports + session digests). Use when the user asks what the memory knows, what is stale, what past sessions taught, or when you need to record a decision, checkpoint or lesson from the shell instead of MCP.

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 0 are distinctive, and 4 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 4

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Every reading command takes --json (field version pins the schema).

    line 16
  2. Very specific:

    Prefer the MCP tools (memory_*) when the server is registered.

    line 35
  3. Very specific:

    Do not edit log.md by hand.

    line 37
  4. Very specific:

    If the CLI is missing: stop and say so (uv tool install memory-boost).

    line 38

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

memory-boost

A router, not a manual. The installed binary's help is the reference: run memory-boost --help once per session and memory-boost <cmd> --help before any subcommand you have not used yet.

Contract

  • stdout = data, stderr = warnings/errors with a fix: line. Exit 0 = trust stdout, 1 = not found / nothing to read, 2 = usage or invalid input. Never retry the same command on exit 2; read stderr and change the input.
  • Every reading command takes --json (field version pins the schema).
  • Reports end with a next: line: run that command before inventing your own.

Routing

NeedCommand
Orient at session start (if the hook did not run)memory-boost brief --cwd .
Find a page or decisionmemory-boost recall "<words>" then memory-boost page <slug>
What in the memory is stale or contradictorymemory-boost drift (or --project <slug>)
How the user's sessions actually go (aggregates only)memory-boost mine --since 7d
Lessons from other projects that apply herememory-boost lessons --project <slug>
Record progress mid-sessionmemory-boost checkpoint --project … --session … --task …
Record a decision or session endmemory-boost save --project … --action …
Record a lesson that travelsmemory-boost lesson --name … --title … --applies-when … --body -

Guardrails

  • mine prints aggregates only; still, never paste transcript contents into the wiki or a report.
  • Prefer the MCP tools (memory_*) when the server is registered; use the CLI for scripts, piping and drift/mine, which have no MCP equivalent beyond memory_drift.
  • Do not edit log.md by hand; save appends it. Pages under wiki/ are yours to edit.
  • If the CLI is missing: stop and say so (uv tool install memory-boost). Do not reimplement it.

Done when

The user's question is answered from wiki content, or the write command printed saved: / checkpoint: / lesson saved:.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free