agentleFS
localoy-ailocaloy-ai/localstack/setup-localbrain/SKILL.md

setup-localbrain

Skill2 starsChanged 41 days agoMITView on GitHub

Install and verify localbrain — the local-first memory store the sales pipeline remembers through; markdown files plus a rebuildable search index, nothing leaves the machine. (localstack)

  • Installs packages

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 1 is distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    No Go: say so plainly and ask (structured question where the runtime supports one) whether to install Go first or skip localbrain.

    line 56
  2. Very specific:

    Never claim installed without command -v localbrain succeeding after.

    line 90

≈ “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

When to invoke this skill

One pass from zero to "localbrain works on this machine": install the CLI, create the store, prove a write/search round trip. Use when asked to "setup localbrain", "install localbrain", or when another skill reported the brain is missing. Re-running is safe — every step detects existing state and repairs only what is missing.

What localbrain is (say this once, in one line)

A local memory store at ~/.localbrain (or $LOCALBRAIN_DIR): markdown files as the source of truth, a derived SQLite search index, no accounts, no cloud. The pipeline uses it to remember leads already contacted or exported, cut reasons and retro learnings across runs. Source: https://github.com/localoy-ai/localbrain

Procedure

1. Detect. command -v localbrain && localbrain version. Already on PATH → skip to step 3.

2. Install. In order of preference, stopping at the first that works:

  • Go toolchain present (command -v go): go install github.com/localoy-ai/localbrain/cmd/localbrain@latest, then confirm it landed: command -v localbrain — if not, the Go bin dir (go env GOPATH/bin) is not on PATH; say exactly that and give the one-line PATH fix for the user's shell rather than working around it.
  • No Go: say so plainly and ask (structured question where the runtime supports one) whether to install Go first or skip localbrain. Every pipeline skill degrades gracefully without it — skipping loses cross-run memory, nothing else. Never fake an install.

3. Init + smoke test. Prove the round trip, then clean up:

localbrain init
echo "setup smoke test $(date +%s)" | localbrain put setup-smoke-test --type note
localbrain search "smoke" | grep -q setup-smoke-test && echo ROUNDTRIP_OK
localbrain delete setup-smoke-test
localbrain doctor

ROUNDTRIP_OK missing or doctor unhealthy → surface the full output and stop. Do not paper over a broken store; the fix is usually localbrain reindex (the index is derived — rebuilding loses nothing).

4. Report. One short block, every row OK/FIX:

localbrain: GREEN
  CLI ........ OK  <version>
  Store ...... OK  <dir> (<N> records)
  Round trip . OK  put -> search -> delete

Then one line on what changes: "/lead-search now dedupes against the brain; /lead-reach records who was contacted and /lead-export what was handed over. The localoy app shows the same records on its Brain screen."

Quality bar

  • Never claim installed without command -v localbrain succeeding after.
  • The smoke record is always deleted — setup leaves no litter in the store.
  • A machine that ends without localbrain ends with an honest one-liner about what the pipeline loses (cross-run memory), not a failure.

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.