agentleFS
AaravChadhaAaravChadha/acstack/skills/contract-check/SKILL.md

contract-check

Skill2 starsChanged 4 months agoMITView on GitHub

Breaking-change pre-flight for the surface callers depend on - function signatures, API response shapes, public exports, config keys. Classifies every change in a diff additive vs destructive, names the add-new-then-deprecate alternative for each destructive one, and opens with a written GO or NO-GO verdict. Never fixes anything - it blocks and explains. Use before merging or releasing a change that touches a public interface, or when the user asks to check for breaking changes.

2 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    Distinctive · estimate

    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
  1. Distinctive:

    Run before the skill's steps — per invocation, not per session (4.36).

    line 22
  2. Very specific:

    Never delete a decision. Supersede it: ~~old~~ → Verdict (YYYY-MM-DD): new call — reason.

    line 40
  3. Distinctive:

    Never fix, tune, or delete a test or eval case to raise a score.

    line 41
  4. Distinctive:

    Pair each destructive item with its alternative: No blank cells; if none exists, say so in the cell.

    line 104
  5. Very specific:

    Every classified item cites file:line from the diff under review.

    line 115
  6. Distinctive:

    Never edits the diff, the code, or the config it is reviewing.

    line 118

≈ “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

/contract-check — the gate before the callers find out

A rename ships, the tests are green, and something downstream breaks that nobody in this repo owns. This skill exists to catch that before the merge. It is structurally read-only: no Edit, no Write, no command that mutates anything. It blocks and explains; it never fixes. (One residual: git log and git diff accept --output=FILE, which a prefix grant cannot exclude; the skill never uses it — see check.sh §13.)

Adjacent skills: /migrate-check (the same gate applied to schema changes; this one covers the code surface) · /ship (runs before release; this gate answers one of its questions in depth).

<!-- acstack:runtime -->

Run before the skill's steps — per invocation, not per session (4.36); failures degrade to markdown:

link="$(readlink "$HOME/.claude/skills/health" 2>/dev/null || true)"   # empty = not symlinked
pack="$(dirname "$(dirname "$link")")"   # NEVER trust this unless $link was non-empty
if [ "${link#/}" != "$link" ] && [ -x "$pack/bin/acstack-config" ] && ! "$pack/bin/acstack-config" runtime | grep -q '=off'; then
  "$pack/bin/acstack-config" || true          # resolved keys, with sources
  "$pack/bin/acstack-update-check" || true    # ≤1 fetch/day; silent ONLY if already checked today
  "$pack/bin/acstack-recall" || true          # LEARNINGS.md + bug-class names, capped 3KB
else
  echo "runtime off — proceeding without recall/update-check"
fi
<!-- /acstack:runtime --><!-- acstack:principles -->

Operating principles

  • Be direct. Push back in writing when the plan or the user is wrong. No sycophancy.
  • Never delete a decision. Supersede it: ~~old~~ → **Verdict (YYYY-MM-DD):** new call — reason.
  • Never fix, tune, or delete a test or eval case to raise a score. Log the miss honestly and leave the case unchanged.
  • Name exact things: regex patterns, function signatures, model names, before → after numbers. Never "fixed bugs".
  • Attribution: follow the project's attribution setting (default none) — no AI-tool mentions in generated docs, no attribution trailers in commits or PRs. Commit with explicit -m/-F messages only.
  • Config: read .claude/acstack.md at the project root (fall back to ~/.claude/acstack.md) before acting. ## Settings keys override pack defaults; a ## <skill-name> section overrides both. Unknown keys and sections are ignored.
  • Docs: BRIEF.md (frozen seed) / PLAN.md (living plan) / JOURNAL.md (rolling journal). If the repo uses legacy names (PLANNING_PROMPT.md / PLANNING.md / STATUS.md), use those instead — never create both.
  • Recall: if LEARNINGS.md exists at the project root, read it before starting.
  • Conduct: follow the acstack-conduct block in this repo's AGENTS.md — the word is the mode; the user sets the pace.
  • Hackathon lane: if the project's AGENTS.md carries the acstack:hackathon-lane block, only /do changes the repository during the event. Any other skill that would write a tracked file, commit or push says what it would have done and stops; a change that is not a task goes through the lane's operator route.
<!-- /acstack:principles -->

The report anatomy is not restated here. It is ../migrate-check/references/gate-shape.md — verdict first, classify every change including the safe ones, a safe alternative per destructive row, never fix, and state what the verdict does not cover. Read it before reporting.

The surface

Four things callers depend on and cannot see you change:

SurfaceWhere to look
Public exportsindex/barrel files, __all__, export statements, package entry points
Function signaturesexported/public functions only — an internal helper is not a contract
API response shapesserializers, response models, schema files, fixtures that encode a response
Config keystemplate/example config, documented keys, environment variable reads

Scope is the diff, not the repo. Everything classified must appear in the range under review. A finding about code the diff never touched belongs in /audit code, not here.

Classification

Decidable rules, not judgment. Each row of the table below is a class, and the third column is what goes in the safe-alternative column.

ChangeClassSafe alternative
Public export renameddestructiveadd the new name, re-export the old, deprecate, remove next major
Public export removeddestructivedeprecate first, remove next major
Field dropped from a response shapedestructiveadd the replacement, mark the old optional, drop next major
Field's type narrowed or nullable → requireddestructiveaccept both, warn on the old form
Required parameter added to a public signaturedestructivegive it a default, or add an overload
Parameter type tighteneddestructiveaccept the union, warn on the narrow case
Config key renamed or removeddestructiveread both keys, warn on the old
Anything added and optional by defaultadditive—
Internal-only change, no exported surface touchedadditive—

A rename is never additive. It looks like one in a diff — a line removed, a line added, the same body — and reporting it as additive is the single failure this skill exists to prevent. When a removed name and an added name share a body, that is a rename until proven otherwise, and the burden is on the evidence that some compatibility shim keeps the old name reachable.

Procedure

  1. Resolve the range. An argument is a ref or a range; with none, use the working diff against the default branch. State which you used — a verdict against the wrong range is worse than no verdict.
  2. Extract the surface. For each of the four surfaces, list what the diff adds, removes and changes. Nothing inferred: cite file:line.
  3. Classify every item, additive or destructive, per the table above.
  4. Pair each destructive item with its alternative. No blank cells; if none exists, say so in the cell.
  5. Verdict. NO-GO if any destructive change lacks an accepted migration path in the diff itself; otherwise GO.
  6. Name the blind spots. Callers outside this repo, runtime consumers no static read reaches, and any surface the range does not include.

Hard rules

  • One verdict per invocation, and it is the report's literal first line — no preamble before it (see gate-shape.md §1).
  • Every classified item cites file:line from the diff under review.
  • A rename reported as additive is a defect in this skill, not a judgment call.
  • Never edits the diff, the code, or the config it is reviewing.
  • GO means "no destructive change without a migration path in this range", never "this change is safe" — say so when the distinction could mislead.

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.