agentleFS
sheldor26sheldor26/whatloads/.claude/skills/recall/SKILL.md

recall

Skill0 starsChanged 57 days agoMITView on GitHub

Pull the relevant slice of project history before writing a plan, and again before the first edit in an unfamiliar or risky area, instead of loading whole log files.

151 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use addskill, apply-firmware and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • ryanportfolioSkill

    addskill

    Create, import, update, or install repo or personal skills: runtime ownership, resources, discovery checks.

    Also in 6 projects that use this

  • ryanportfolioSkill

    apply-firmware

    Bring current folder (plain folder, existing repo, old or partial firmware copy) up to full Harness Firmware in place. Mirror external repo → adopt-repo.

    Also in 6 projects that use this

  • +10 more from ryanportfolio

The file

Files
SKILL.md

Recall before you plan

The logbook is deliberately not loaded into context automatically. Reading MISTAKES.md in full costs thousands of tokens for history that is irrelevant to the task at hand. Retrieve by tag instead.

When to do this

Before writing the plan, not before the first edit. This is the part people get wrong, including the first version of this skill.

A plan built without the history bakes the repeat mistake into the plan itself, and then implementation carries it out faithfully. By the time the first edit comes around, recalling the entry means either discarding the plan or quietly routing around the guardrail — and routing around it is what actually happens. Two minutes earlier, the same entry changes the design for free.

So: in plan mode, recall is the first move, before the plan is drafted. Then again before the first edit in any area not touched this session. Always for auth, payments, migrations, data writes, scrapers, deploy config, and anything that touches money or user data.

How

node .bitacora/cli.mjs recall <tag>          # by area:      recall pricing
node .bitacora/cli.mjs recall <path>         # by file:      recall scripts/import.ts
node .bitacora/cli.mjs recall <phrase>       # full text:    recall "race condition"
node .bitacora/cli.mjs stats                 # where friction concentrates

recall prints ids, titles and tags — a cheap index. Then open the file and read the full text of the two or three entries that actually apply. Do not skim; the guardrail is usually the last line of the entry and it is the part that matters.

What to do with what you find

  • A mistake entry with a guardrail: honour the guardrail. If it says a script must not overwrite manually verified values, do not write code that can. If the guardrail looks wrong, say so — do not route around it silently.
  • A mistake entry without a guardrail: that entry is unfinished. Propose the guardrail as part of this task.
  • A decision entry: check whether its Context still holds. If the context has changed, that is worth raising explicitly rather than quietly doing the opposite.

If recall finds nothing

Say so, and proceed carefully. An area with no history is an area where the first mistake has not been made yet — which makes it more likely, not less, that this session makes it.

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.