Rule catalog
The rules the fleetlint:* presets ship, grouped by family, with a one-line summary of what each checks. The rule reference has a page per rule, generated from the catalog, with the exact requirement, parameters and fix. The practices behind the rules are in the baseline and, per stack, in the stack profiles.
Reading the tables
Preset names the smallest preset that selects the rule:
minimal: the floor for any repository. Nothing secret, accidental or oversized is tracked, no conflict markers are committed, a public repository has a license, and hooks exist. The two hook rules apply at tiers 1 and 2, not to experiments.recommended, the default: minimal plus the families below.strict: recommended with every severity raised one step. The security rules take no exceptions and cannot be disabled.oss: recommended with its public/* rules run whatever the detected visibility.slop: an add-on with the heuristic slop/* rules and the two agent-trace rules for public repositories.
Severity is the default. A repository may raise it freely and lower it with a reason.
Tiers: all, or the tiers the rule applies to. What runs at each tier lists them per tier.
Kind says how the rule decides:
expr: a CEL predicate. A repository can except a finding, but cannot add ways to pass.outcome: an outcome rule. It names a result and lists the known ways of reaching it, its satisfiers; the first that holds passes the rule and is reported as the evidence. A repository adds its own way with accept: (writing rules).capability: one requirement that each stack meets with its own tools. lint/configured is implemented by lint/configured/go, lint/configured/python and so on, each an expr or outcome rule with its own page. Every implementation that runs must pass. Selecting, overriding or excepting the capability reaches every implementation, and accept: reaches its outcome implementations; an implementation id adjusts one stack. The capability matrix shows which stack implements which (writing rules).capability by artifact: a capability implemented per kind of published artifact instead of per stack. Each image, package or set of release files is checked on its own (artifacts).go: written in Go inside fleetlint. Only quality/check-passes, which runs the repository’s code.
repo: repository hygiene
| Rule | Severity | Tiers | Preset | Kind | Checks |
|---|
repo/no-tracked-junk | error | all | minimal | expr | no OS, editor or backup files, bytecode or dependency directories tracked (.DS_Store, *.log, __pycache__/, node_modules/, …) |
repo/no-tracked-env | error | all | minimal | expr | no .env* with values tracked (.env.example allowed) |
repo/no-large-files | warning | all | minimal | expr | no tracked file over params.max_kb (default 2048) unless git-lfs takes its pattern |
repo/gitignore-present | warning | all | minimal | expr | a .gitignore at the root |
repo/editorconfig | warning | all | recommended | expr | an .editorconfig exists |
repo/gitattributes | warning | all | recommended | expr | .gitattributes contains text=auto |
repo/readme-present | warning | all | recommended | expr | a README of at least 10 lines with a heading on installing, running or using the project |
repo/lockfile-committed | error | 1, 2 | recommended | capability | the lockfile of each stack’s dependency manager is tracked |
repo/toolchain-pinned | warning | 1, 2 | recommended | capability | each stack pins its compiler or runtime with a version (the stack’s version file, .tool-versions/mise.toml naming the tool, or devbox.json naming its package with a version); a JDK pin for Java and Kotlin |
repo/dev-environment | warning | 1, 2 | recommended | outcome | the tools are reproducible: in a container mode the container; on the machine a Nix flake, devenv or devbox with its lock file (or shell.nix), mise.toml / .tool-versions, or a tools target with pinned versions |
repo/codeowners | info | 1 | recommended | expr | a CODEOWNERS file at the root or in .github/, docs/ or .gitea/ |
taskrunner: the task-runner contract
| Rule | Severity | Tiers | Preset | Kind | Checks |
|---|
taskrunner/targets | error | 1, 2 | recommended | expr | fmt lint test cover audit check check-fast exist and are not placeholders (Makefile, justfile, Taskfile or package.json scripts; not checked with Gradle, whose check is the convention) |
taskrunner/container | warning | 1, 2 | recommended | outcome | when facts.container is not none: a target called outside the container runs inside it (devcontainer CLI, docker, podman, or an accept:ed way); Makefile, justfile, Taskfile |
taskrunner/check-composition | warning | all | recommended | expr | check reaches lint and test (up to three levels deep) and neither bench nor release; Makefile, justfile, Taskfile |
The per-stack tools and settings are in the stack profiles.
| Rule | Severity | Tiers | Preset | Kind | Checks |
|---|
format/configured | warning | 1, 2 | recommended | capability | a formatter is configured or runs (gofumpt, ruff format, rustfmt, Prettier or Biome, Spotless, dart format, dotnet format, clang-format, RuboCop, PHP CS Fixer, …); not applicable to Hugo |
lint/configured | warning | 1, 2 | recommended | capability | a linter is configured (golangci-lint v2, ruff, clippy [lints], ESLint/Biome/oxlint, detekt 2, Checkstyle/Error Prone/SpotBugs/PMD, .NET analyzers, clang-tidy, RuboCop, PHPStan/Psalm, a Dart lint set; for Hugo hugo --panicOnWarning) |
lint/strict | warning | 1, 2 | recommended | capability | the linter’s required checks are on, one finding per missing check (golangci-lint errcheck govet staticcheck errorlint gosec gocognit funlen; ruff E F I B UP SIM BLE S C90 T20; clippy unwrap_used expect_used panic todo dbg_macro; five detekt rules; .NET Nullable, warnings as errors, analysis level; clang-tidy WarningsAsErrors; Java analyzers that fail the build; RuboCop NewCops: enable) |
types/strict | warning | 1, 2 | recommended | capability | the type checker runs in strict mode (tsconfig strict flags; mypy or pyright strict; PHPStan max or Psalm level 1; the Dart analyzer’s three strict modes); not applicable where the compiler is the type checker |
lint/shell | info | all | recommended | expr | when shell scripts are tracked, shellcheck runs as a hook, in the task runner or in CI |
quality: behaviour, not configuration
| Rule | Severity | Tiers | Preset | Kind | Checks |
|---|
quality/check-passes | error | 1, 2 | recommended | go | with --deep: the task runner’s check exits 0 within the budget |
quality/policy-enforced | warning | 1, 2 | recommended | expr | fleetlint check runs as a hook, in the task runner or in CI |
quality/coverage-threshold | warning | 1, 2 | recommended | capability | tests fail below a coverage floor (the stack’s coverage tool with a minimum, diff-cover --fail-under, a codecov target) |
quality/duplication-gate | info | 1 | recommended | capability | a duplication detector runs: golangci-lint dupl, pylint duplicate-code, PMD CPD, or for every stack jscpd, Simian, duplo or Sonar |
quality/mutation-gate | info | 1 | recommended | capability | a mutation testing tool runs on the core: gremlins, mutmut, cargo-mutants, StrykerJS, Pitest, mutation_test, Stryker.NET, Mull, mutant, Infection |
docs and changelog: the documents that must exist
| Rule | Severity | Tiers | Preset | Kind | Checks |
|---|
docs/architecture | warning | 1 | recommended | expr | docs/ARCHITECTURE.md, ARCHITECTURE.md or an Architecture section in the README |
docs/adr-directory | info | 1 | recommended | expr | at least one Markdown record in docs/adr/ (or docs/decisions/, doc/adr/, adr/, …) |
changelog/present | warning | 1, 2 | recommended | expr | CHANGELOG.md with at least one ## section |
changelog/generator-configured | info | 1 | recommended | outcome | a changelog generator is configured: git-cliff, GoReleaser, release-please, cocogitto, semantic-release, commitizen, changesets or towncrier |
hooks: prek, pre-commit or lefthook
| Rule | Severity | Tiers | Preset | Kind | Checks |
|---|
hooks/config-present | error | 1, 2 | minimal | expr | .pre-commit-config.yaml or a lefthook configuration |
hooks/secret-scan | error | 1, 2 | minimal | outcome | a pre-commit hook runs a secret scanner (gitleaks, trufflehog, ggshield, detect-secrets, secretlint); detect-private-key alone is a partial result |
hooks/shared-hygiene | warning | 1, 2 | recommended | expr | .pre-commit-config.yaml runs the shared hygiene hooks (trailing whitespace, end of file, merge conflicts, large files); lefthook is not read |
hooks/conventional-commits | warning | 1, 2 | recommended | outcome | a commit-msg hook checks every message |
hooks/pre-push-check | warning | 1, 2 | recommended | outcome | a pre-push hook runs the task runner’s check (check-fast does not count) |
hooks/revs-pinned | warning | 1, 2 | recommended | expr | every rev: in .pre-commit-config.yaml is a version tag or a commit SHA, never a branch |
deps: dependency hygiene
| Rule | Severity | Tiers | Preset | Kind | Checks |
|---|
deps/locked-install | warning | 1, 2 | recommended | capability | installs use the lockfile strictly (npm ci, uv sync --locked, cargo --locked, composer install, Bundler frozen, …) and nothing rewrites it outside an update target |
deps/vulnerability-scan | warning | 1, 2 | recommended | capability | the task runner or CI runs a vulnerability scanner (the stack’s tool, or osv-scanner, trivy, grype and the like for every stack) |
deps/license-check | info | 1 | recommended | capability | the task runner or CI checks dependency licenses against an allowlist |
deps/unused-check | info | 1 | recommended | capability | unused dependencies are detected (go mod tidy -diff, deptry, cargo-machete, knip, composer-unused, dependency_validator, …) |
deps/update-automation | warning | 1, 2 | recommended | expr | Renovate or Dependabot is configured |
deps/update-coverage | info | 1, 2 | recommended | expr | the update automation covers every detected stack, GitHub Actions and Dockerfiles |
deps/update-cooldown | info | 1, 2 | recommended | expr | updates wait a few days after a release: Renovate minimumReleaseAge, Dependabot cooldown |
deps/install-scripts-disabled | info | 1, 2 | recommended | expr | Node: install scripts are off or allowlisted (.npmrc, .yarnrc.yml, pnpm onlyBuiltDependencies, bun trustedDependencies) |
deps/forks-documented | info | 1, 2 | recommended | capability | a dependency taken from a fork (go.mod replace, git sources in Cargo.toml, package.json, Python requirements or the Gemfile, a VCS repository in composer.json) is documented in FORKS.md, a Forks section, or a comment on its line |
containers: repositories with a Dockerfile
| Rule | Severity | Tiers | Preset | Kind | Checks |
|---|
containers/dockerignore | warning | all | recommended | expr | a .dockerignore exists |
containers/non-root-user | warning | all | recommended | expr | every Dockerfile sets a non-root USER or builds on a non-root base (:nonroot, Chainguard); one finding per file |
containers/base-image-pinned | info | 1 | recommended | expr | every FROM carries @sha256:; scratch, build arguments and earlier stages exempt |
ci: the check workflow and its safety
The CI rules read GitHub, Gitea and Forgejo Actions workflows and .gitlab-ci.yml with its local includes. A rule about something only one system has applies to that system only; the rule’s page says which.
| Rule | Severity | Tiers | Preset | Kind | Checks |
|---|
ci/check-workflow | warning | 1, 2 | recommended | outcome | a workflow or pipeline on push or pull/merge request runs the task runner’s check (check-fast does not count) |
ci/actions-pinned | warning | 1 | recommended | expr | every uses: is pinned to a commit SHA, docker:// images by digest; GitLab includes and components to a commit SHA, remote includes by integrity: |
ci/least-privilege | warning | 1, 2 | recommended | expr | GitHub Actions only: top-level or per-job permissions: present and not write-all |
ci/no-script-injection | warning | 1, 2 | recommended | expr | Actions: no ${{ }} over outsider-controlled event data inside a run: or github-script script: |
ci/no-pipe-to-shell | warning | 1, 2 | recommended | expr | no download piped into a shell in CI scripts or Dockerfiles |
ci/no-pull-request-target | error | all | recommended | expr | no pull_request_target workflow checks out or fetches the PR’s code (GitHub, Gitea, Forgejo) |
ci/checkout-no-persist-credentials | info | 1 | recommended | expr | Actions: every actions/checkout sets persist-credentials: false |
ci/workflow-lint | info | 1 | recommended | outcome | Actions workflows are linted with actionlint or zizmor |
ci/sast | info | 1 | recommended | capability | a SAST tool runs: CodeQL, Semgrep or Opengrep, GitLab SAST, SonarQube; for Ruby also Brakeman, for PHP Psalm’s taint analysis |
ci/job-timeouts | info | all | recommended | expr | every job sets a timeout (timeout-minutes; GitLab timeout:); one finding per job |
ci/concurrency-cancel | info | all | recommended | expr | pull-request pipelines cancel superseded runs (concurrency with cancel-in-progress; GitLab interruptible) |
release: producing artifacts
The release rules apply once a repository releases: a release mechanism, a tag-triggered pipeline, a published artifact or version tags. Each rule’s page says which.
| Rule | Severity | Tiers | Preset | Kind | Checks |
|---|
release/tag-triggered | error | 1 | recommended | expr | releases come from a tag-triggered pipeline (Actions push: tags:, GitLab $CI_COMMIT_TAG) |
release/signing | error | 1 | recommended | capability by artifact | every published artifact is signed: with its ecosystem’s signature where there is one (APK signing, PGP for Maven Central, npm provenance, PyPI attestations, NuGet signing), otherwise with cosign or gpg |
release/sbom | warning | 1 | recommended | capability by artifact | an SPDX or CycloneDX SBOM is generated in the release pipeline for each artifact that carries its own dependencies; packages a registry resolves from a manifest (npm, PyPI, crates.io, RubyGems, pub.dev) are not checked |
release/provenance | warning | 1 | recommended | capability by artifact | build provenance is attested: by the registry where it does that (npm, PyPI, RubyGems), otherwise attest-build-provenance, the SLSA generator, cosign attest, BuildKit or GitLab artifact metadata |
release/checksums | warning | 1 | recommended | outcome | a checksums file is produced (GoReleaser default, sha256sum, shasum -a 256); not needed for what a registry addresses by digest |
release/trusted-publishing | warning | 1 | recommended | capability by artifact | publishing to npm, PyPI, crates.io, RubyGems, NuGet or pub.dev uses OIDC trusted publishing, not a registry token from secrets (GitHub and GitLab) |
release/semver-tags | warning | all | recommended | expr | existing tags match v?MAJOR.MINOR.PATCH[-pre]; one finding per offending tag |
release/signed-tags | warning | 1 | recommended | expr | the newest params.recent (3) version tags carry a PGP, SSH or X.509 signature |
release/changelog-in-release | info | 1 | recommended | outcome | release notes come from git-cliff, GoReleaser, release-please, changesets or the generated CHANGELOG.md, not hand-written |
release/hosted-builder | info | 1 | recommended | expr | jobs of a tag-triggered GitHub workflow run on GitHub-hosted runners |
release/reproducible-build | info | 1 | recommended | capability by artifact | release builds are reproducible: the stack’s build settings (-trimpath, --remap-path-prefix, outputTimestamp, ContinuousIntegrationBuild) or SOURCE_DATE_EPOCH in the release build |
release/api-compatibility | info | 1 | recommended | capability | libraries: the stack’s API check runs against the previous release (gorelease or apidiff, griffe, cargo-semver-checks, API Extractor, binary-compatibility-validator, japicmp or revapi, .NET package validation, dart_apitool) |
public: open-source repositories (when: repo.public)
| Rule | Severity | Tiers | Preset | Kind | Checks |
|---|
public/license | error | all | minimal | expr | a license file at the root (LICENSE*, LICENCE*, COPYING*, UNLICENSE) or a REUSE LICENSES/ directory |
public/security-policy | warning | 1 | recommended | expr | SECURITY.md at the root or in .github/, docs/ or .gitea/ |
public/contributing | info | 1 | recommended | expr | CONTRIBUTING.md at the root or in .github/, docs/ or .gitea/ |
public/code-of-conduct | info | 1 | recommended | expr | CODE_OF_CONDUCT.md at the root or in .github/, docs/ or .gitea/ |
public/manifest-license | info | 1 | recommended | capability | a published package’s manifest declares its license (package.json, pyproject.toml, Cargo.toml, the Maven POM, a gemspec, composer.json, PackageLicenseExpression, vcpkg.json, conanfile.py) |
public/issue-templates | info | 1 | recommended | expr | an issue or pull-request template in .github/ or .gitea/ |
public/no-agent-files | error | all | slop | expr | no agent instruction files tracked (CLAUDE.md, CLAUDE.local.md, AGENTS.md, GEMINI.md, .github/copilot-instructions.md, .cursorrules, .windsurfrules, .clinerules, .aider*, .claude/, .agent/, .cursor/) |
public/no-ai-attribution | error | all | slop | expr | the last params.commits (default 200) commits carry no AI tool’s address as author or co-author and no generated-with footer; names are not read |
slop: traces of careless or generated code
Catalog and rationale: slop.md. slop/conflict-markers is in fleetlint:minimal and three more are in fleetlint:recommended; the others are heuristics in the add-on preset fleetlint:slop.
| Rule | Severity | Tiers | Preset | Kind | Checks |
|---|
slop/conflict-markers | error | all | minimal | expr | no merge conflict markers in source |
slop/focused-tests | error | all | recommended | expr | no .only, fit, fdescribe, solo: true, RSpec focus: true |
slop/scratch-scripts | warning | all | recommended | expr | no fix_*, debug-*, tmp_* scripts at the root |
slop/report-documents | warning | all | recommended | expr | no *_SUMMARY.md, *REPORT*.md, NEXT_STEPS.md and the like |
slop/narration-comments, slop/placeholder-comments, slop/banner-comments, slop/todo-without-issue, slop/commented-out-code | info | all | slop | expr | comment hygiene |
slop/emoji-in-code, slop/typographic-chars, slop/emoji-headings | info | all | slop | expr | no emoji or typographic characters in code and headings |
slop/marketing-words | info | all | slop | expr | no marketing filler in Markdown documentation |
slop/debug-output, slop/arbitrary-delays, slop/swallowed-errors, slop/escape-hatches, slop/forced-unwraps, slop/vague-errors, slop/placeholder-values | info | all | slop | expr | code smells outside tests |
slop/insecure-patterns | warning | all | slop | expr | no disabled TLS verification, eval(, chmod 777, wildcard CORS |
slop/variant-names, slop/variant-files, slop/dumping-ground-files | info | all | slop | expr | no V2/New/Old variants, no utils dumping grounds |
slop/skipped-tests, slop/trivial-asserts | info | all | slop | expr | test hygiene |
slop/vague-commit-messages, slop/non-conventional-commits, slop/huge-commits | info | all | slop | expr | history of the last params.commits commits |
Totals
| Count |
|---|
| rules | 106 |
| of which capabilities | 22 |
| of which keyed by artifact | 5 |
implementations per stack (and any) | 170 |
| implementations per artifact kind | 65 |
Templates are not rules
The rules check outcomes, not the shape of the files fleetlint fix writes: a repository with its own hooks, workflows or release pipeline passes when it produces the same results. fleetlint’s own tests check the other direction: a repository set up from each stack’s templates passes the hook, CI and release rules.