agentleFS
M-BajalanM-Bajalan/kibsu-lab/skills/gate/SKILL.md

gate

Skill0 starsChanged 31 days agoMITView on GitHub

What may not happen without a green gate in this repository. Names the reversibility rule, the commands that check it, and the exact exit codes that mean pass/fail/could-not-run.

32 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use new, build 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.
  • codebycoreySkill

    new

    Start a new change with a proposal. Use when starting a feature, creating a proposal, or beginning planned work.

    Also in 3 projects that use this

  • Microsoft AzureSkill

    build

    Build the TypeSpec Rust emitter from source. Run this after cloning the repo, after pulling changes, or when TypeScript source files have been modified.

    Also in 5 projects that use this

  • AWSSkill

    ship

    Finalize dev work on an issue branch — update docs, push, and open a PR.

    Also in 8 projects that use this

  • genomewalkerSkill

    checkpoint

    Save a checkpoint of task status, decisions, active files, and next steps

    Also in 3 projects that use this

  • VercelSkill

    investigate

    Investigate an issue report, or analyze the constraints of a proposed change, with strict, evidence-driven diagnosis.

    Also in 3 projects that use this

  • BlockSkill

    plan

    Draft a TDD, PRD, or lightweight plan in the conversation without committing it.

    Also in 4 projects that use this

The file

We split this skill into the rules it gives an agent

4 rules. 0 are common, 2 are distinctive, and 2 are very specific to this project.

  • 0

    Common

    Most of the 30 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 2

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  • could be automatic: a setting could enforce it
  1. Very specific:

    Read .kibsu.json directly to see which commands are configured before assuming none are.

    line 29
  2. Very specific:

    Before reporting any of T1/T2/T3 as done:

    line 58
  3. Distinctive:

    No irreversible action -- an overwrite with no snapshot, a force-push, a delete with no undo path -- without it being explicitly, …

    line 72
  4. Distinctive:

    After any change, re-run the check that names the invariant -- never just re-read the diff you produced and conclude it worked.

    line 78

“Common” is measured: kept by at least 6 in 10 of the 30 edited versions of this file we could read. “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

gate

The rule

Make a change reversible if you can. Gate it only if you can't. Before any change, ask: can this be cheaply taken back? If yes, act and report -- no ceremony. If no, name what you'll verify afterward, in one line, before you touch anything.

What this repo already gates, mechanically

python -m kibsu gate --check runs every command configured under .kibsu.json's "gates" key and blocks a commit only on a NEW violation -- never on the pre-existing baseline. Check what's wired and whether it's live:

python -m kibsu gate --status

Read .kibsu.json directly to see which commands are configured before assuming none are. A repo with "gates": [] has zero enforcement and kibsu gate --check says so plainly -- that is an abstention, not a pass.

Before any commit

python tools/reconcile.py

Exit 0 = clean. Exit 1 = a real reconciliation mismatch -- do not commit past it silently; either it's a pre-existing accepted baseline violation or it's new and must be fixed first. Exit 2 = could not run (data missing and could not be generated) -- this is tools/reconcile.py's documented cannot_run_exit; a gate wired with that value skips the check rather than blocking, and says so on stderr.

To tell a pre-existing mismatch from a new one:

python -m kibsu gate --check

Exit 0 = everything tools/reconcile.py reports is already accepted in .kibsu/gate_baseline.json. Exit 1 = something is new, and the output names it. Never run python -m kibsu gate --baseline to settle the question: it does not show the baseline, it rewrites it to accept whatever is wrong right now.

Task checkers are gates too

Before reporting any of T1/T2/T3 as done:

python tools/check_task.py T1
python tools/check_task.py T2
python tools/check_task.py T3

Exit 0 = verified done. Exit 1 = not done, or a claim of "done" that the checker could not confirm from the repo -- treat exit 1 the same as a failing test, never as a formality to route around. Exit 2 = the checker itself could not run.

The one hard rule

No irreversible action -- an overwrite with no snapshot, a force-push, a delete with no undo path -- without it being explicitly, just-in-time approved for exactly that action. A plan is not approval. Presenting options is not approval.

Verify from the source, not from your own report

After any change, re-run the check that names the invariant -- never just re-read the diff you produced and conclude it worked. git status and git diff show what changed; they do not prove the change is correct. The commands above do.

Checklist before you say "done"

  • Is the change reversible? If yes, did you act and report rather than seeking permission you didn't need?
  • If irreversible, did you name the undo path (or state plainly "no undo exists") before acting?
  • Did you run the relevant checker (tools/reconcile.py, tools/check_task.py) and read its exit code, not just its printed text?
  • Did you log the change per skills/change-log/SKILL.md?

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.