Menu
Interprets ironlint PreToolUse hook output after Edit/Write -- fixes the blocked edits it reports.
Installer avec Codex ou Claude Copiez ce prompt, collez-le dans Codex, Claude ou un autre assistant, puis laissez-le vérifier la page du skill et l'installer pour vous.
Basé sur la classification professionnelle SOC
Bootstraps a project's .ironlint.yml with IronLint's generic baseline, then helps the user deliberately add checks for their stack and existing linters. Use when user says "init ironlint", "set up ironlint", "bootstrap ironlint config", "create ironlint checks", "ironlint init", or asks to create or generate an ironlint configuration.
Authors, modifies, or removes checks in an ironlint .ironlint.yml. Use when the user says "add an ironlint check for X", "ban Y", "tighten <check-id>", "stop checking <check-id>", "remove <check-id>", "change the scope of a check", or asks how to write an ironlint config.
Use when checking whether an IronLint adapter still matches its coding harness's current contract — auditing adapter/harness drift, verifying hook payload shapes, plugin manifest schemas, lifecycle events, or tool names are up to date, or doing periodic adapter maintenance. Takes a harness name (claude-code, codex, pi, opencode) as argument.
Reviews ironlint check health from the telemetry log. Use when the user says "review my ironlint checks", "check health", "which ironlint checks are noisy", "find dead ironlint checks", "ironlint review", or asks for an audit of .ironlint.yml.
Removes transient build artifacts this task produced in the ironlint repo before declaring work done. Use at the end of any task that ran `cargo build --release`, `cargo mutants`, `cargo llvm-cov`, generated a one-off binary, wrote a scratch file like `pr.diff` or an ad-hoc tarball, or any other throwaway output. Do NOT use to wipe the actively-iterating `target/` directory.
| name | ironlint |
| description | Interprets ironlint PreToolUse hook output after Edit/Write -- fixes the blocked edits it reports. |
| metadata | {"author":"dynamik-dev","version":"2.1.0","category":"workflow-automation","tags":["linting","hooks","code-quality","pre-tool-use"]} |
Interpret and act on ironlint PreToolUse hook output. Not user-invocable.
The tool-result stderr is a Verdict JSON whose status is block. A check is a
shell command; it blocked because it exited nonzero (1–125). The adapter uses
hook exit 2 to reject the tool call. Shape:
{
"schema_version": 6,
"status": "block",
"blocks": [
{"check": "no-debug", "step": null, "file": "src/foo.rs", "message": "src/foo.rs:42: DEBUG marker"}
],
"errors": [],
"passed": ["no-todo"]
}
Each entry in blocks is one check that rejected the edit:
check — the check id that blocked (defined in .ironlint.yml).file — the file it checked.message — the check's own combined output, verbatim. This is your instruction
for what to fix; if the check emits line numbers (e.g. a grep -n or a linter),
they're in here.Fix every entry in blocks in the named file before any other tool call. The
hook re-fires on the next Edit and re-checks. Repeat until blocks is empty.
passed lists the check ids that ran and passed for this file. Their concerns are
already satisfied — don't re-investigate them.
errors lists checks that couldn't run (not found, timed out, or killed) — each
is {check, file, reason}, not a policy violation. By default the hook fails open
on these (the edit is allowed and nothing reaches you); you'll only see an error
note when the project set IRONLINT_FAIL_CLOSED_ON_INTERNAL=1. A check that couldn't
run is a broken check, not a finding — surface it, don't try to satisfy it.