sarj-lint-configs

Ships the maximally-strict ruff / pyright / ESLint configs from sarj-ai/standards as a pip-installable package, plus the commands that adopt them and keep them current.

Adopt

uvx --from sarj-lint-configs sarj-standards init

init detects Python and/or TypeScript, installs the toolchain and hooks, and writes only the configs that ecosystem uses, wires them up ([tool.ruff] extend, pyrightconfig.json, eslint.config.mjs), writes a pre-commit block, records the adopted version in .sarj-standards.toml, and prints the CI snippet and — for TypeScript — the one ESLint peer set that resolves together. Pass --no-install to write only the wiring.

init --dry-run prints the whole plan without touching anything. Existing Ruff, Pyright, ESLint flat-config, package-manager, and pre-commit files are merged only when their structure is unambiguous; user-owned settings are preserved. If a file cannot be wired safely, init explains the exact manual change and makes no changes. A failed sync or dependency install restores every tracked file.

Application repositories can opt into the preferred-library policy at adoption time:

uvx sarj-lint-configs init --profile application

The selected standard or application profile is recorded in .sarj-standards.toml, so later update --configs-only, update --configs-only --check, and check select the same standalone Ruff and ESLint artifacts automatically. Existing manifests without a profile field remain on standard. The application profile is an intentional stack policy: it treats cataloged imports such as argparse and pandas as errors even when the old library remains maintained.

It also checks direct dependencies in pyproject.toml, PEP 735 groups, Poetry/PDM/uv tables, authored requirements files, and every npm dependency field. Run that manifest gate independently—or measure an unadopted repository— with:

uvx sarj-lint-configs check --dependencies --profile application

Findings carry stable LIB### IDs and migration cautions. Architectural replacements such as Flask to FastAPI, pandas to Polars, and Axios to Ky are deliberately errors, but are never broadly autofixed because their APIs and runtime semantics differ.

Your TypeScript does not have to be at the repo root

init writes each ecosystem's configs into the directory that owns it, not into the repo root: the ESLint config goes beside the npm lockfile, the ruff and pyright configs beside the pyproject.toml. That is not cosmetic. A repo whose TypeScript lives in frontend/ has no node_modules and no tsconfig.json at its root, and ESLint does not search upward for a flat config — a root-level eslint.config.mjs is a file that can never load, written by a tool that reports success.

The project root is found by lockfile, not by package.json: a repo can carry a root package.json declaring nothing but packageManager while the real project is a directory down.

The detected destinations are recorded in .sarj-standards.toml, so plain update --configs-only and update --configs-only --check find them again without CI having to restate them. Override detection with init --python-dest / --typescript-dest.

npm, pnpm, Yarn and Bun

The peer set below does not install without an override block, and every client spells that block differently: npm and Bun use package.json#overrides, pnpm 11 wants overrides in pnpm-workspace.yaml (older standalone pnpm projects use package.json#pnpm.overrides) with a parent>child selector, Yarn wants resolutions with a parent/child path and no $dep indirection. init detects the client from the lockfile (or a packageManager field, which Corepack enforces), writes the right one, and prints the matching install command. Writing npm's spelling everywhere was worse than writing nothing: pnpm and Yarn ignore a stray overrides key, so the install failed identically while package.json looked fixed.

Keep current

uv run --frozen sarj-standards doctor       # read-only diagnosis with exact fixes
uv run --frozen sarj-standards update --check   # preview whether an upgrade is needed
uv run --frozen sarj-standards update       # resolve latest, migrate, install, postflight
uv run --frozen sarj-standards check        # run every adoption and lint gate
uv run --frozen sarj-standards fix          # apply formatter and safe lint fixes
uv run --frozen sarj-standards show state   # print detected adoption state
uv run --frozen sarj-standards check --noise-only .  # comment/artifact ratchet only

Run check in CI. init prints it as a ready-made job. With no paths, check runs the complete repository verification. With paths, it runs the applicable custom rules for those paths, which keeps pre-commit fast. check --noise-only covers Python and text/config inputs; TypeScript comment noise is enforced by the generated strict ESLint config.

check checks the whole repository by default. A tool repository whose rule fixtures intentionally contain rejected code can narrow only the custom-rule pass, without weakening Ruff, BasedPyright, or ESLint:

[verify]
paths = ["src", "tests", "README.md"]

Paths are repository-relative and cannot escape the repository root.

Maintainers can declare repository policy in .sarj-standards.toml and replace ad hoc scripts with sarj-standards maintain check. The same command checks private references, CI history, filenames, rule/test/registry pairing, canonical config references, and version coverage; maintain sync-ledger, maintain comment-corpus, maintain hooks install, and show rules provide the related maintenance operations.

The package also owns repository setup and release policy; workflows and the Makefile are intentionally thin adapters rather than a second implementation:

sarj-standards maintain setup --check
sarj-standards maintain release check-tag typescript-v9.13.0
sarj-standards maintain release lock-age packages/typescript/package-lock.json --exclude-file .github/release-age-exclusions.txt
sarj-standards maintain release typescript check
sarj-standards maintain release publish typescript

The former sync, list, path, peers, upgrade, verify, format, inspect, library-policy, and repo commands remain exact compatibility aliases. Existing CI and hooks keep working while new usage stays within the seven consumer verbs shown by sarj-standards --help.

Programmatic consumers should import sarj_lint_configs.api. Business logic is grouped under sarj_lint_configs.libs by adoption, linting, repository, setup, and release domains; __main__ is only an entrypoint and cli/ owns argument parsing and presentation. The former top-level modules remain identity-preserving compatibility aliases for existing imports and monkeypatches.

The Python facade mirrors the consumer vocabulary without parsing CLI output:

from pathlib import Path

from sarj_lint_configs.api import Standards

standards = Standards(Path.cwd())
health = standards.doctor()
preview = standards.init(dry_run=True)
result = standards.init(install=False)
status = standards.check(("src", "tests"))

The preferred API is deliberately small: Standards, Result, Finding, Change, Inspection, Status, and __version__. Each method calls the same typed services as the CLI. Existing sarj_lint_configs.api exports remain as compatibility aliases. Maintainer plan/apply, rule evaluation, corpus, and release APIs live in the public sarj_lint_configs.libs namespace.

Release-age exception files contain one exact package@version per line, with optional # comments. Name-wide exceptions are rejected in files, so a future lockfile upgrade cannot silently inherit a temporary whitelist.

sarj-standards is the preferred entrypoint. sarj-lint-configs remains an alias for compatibility. check routes Python, SQL, Terraform, YAML/TFTPL, TOML, JSONC, Markdown, INI-style config, environment files, shell scripts, Dockerfile variants, Makefiles, and Justfiles in one pass. It rejects commented-out config (SARJ301), dense config narration (SARJ300), and AI execution diaries, bug-hunt dumps, or large point-in-time audits that should have been reduced to durable README/docs/ADR facts (SARJ302). Large documents need at least two independent artifact signals, so size alone never reports. SARJ302 is visible but non-blocking for its first minor release; promotion is an explicit per-rule metadata change after corpus calibration, never a global warning-to-error rewrite. The new constructor and shadcn checks are likewise warnings, and shadcn guidance exists only in the application profile, outside tests, fixtures, and design-system implementation directories.

Generated or intentionally instructional config can be excluded explicitly; there is no blanket directory exemption:

[text]
exclude = ["examples/generated-values/**"]

In source code, use names, types, small functions, and data structures to carry the explanation. One sentence of nearby rationale is clean; exactly two sentences is a non-blocking warning; three or more is an error and should be reduced to the local constraint or moved to an ADR/doc. Fully typed functions must not repeat parameters or returns in docstring/JSDoc tables, while external contracts, invariants, failures, examples, generated docs, directives, licenses, and runtime-consumed prompt/tool/route documentation remain valid.

doctor — one version, not three

Run it without preparation:

sarj-standards doctor
sarj-standards doctor --format json  # stable schema and finding IDs for automation

It is read-only, exits 0 for a healthy or not-yet-adopted repository, 1 for actionable drift, and 2 for invalid input such as malformed TOML/JSON. Every drift finding includes a concrete remediation. Intentional compatibility fixtures can be excluded narrowly with [doctor] exclude = ["tests/fixtures/**"].

Upgrading is one command from an installed environment:

sarj-standards update

The command bootstraps the newest published compatibility bundle with uvx --refresh, previews each changed path, blocks on retired rule references, updates the single manifest version, syncs configs, safely repairs wiring and peer pins, installs dependencies, and rolls everything back if installation or the doctor postflight fails. Use update --check in automation. The legacy upgrade --offline --no-install spelling remains available for maintainers testing the already-installed bundle.

A consumer repo used to state a Sarj version in three independent places: the pyproject.toml pin, the pre-commit rev:, and whatever a CI job typed on its own command line. Nothing compared them, so they drifted apart and stayed drifted — a repo could run one linter version at commit time, a second in CI, and a third locally, and pass its own build.

doctor finds every pin site under a repo root and checks each against the installed wheel:

ok     .sarj-standards.toml  --  version 0.32.0
drift  .github/workflows/ci.yml: sarj-python-lint==0.12.2  --  installed sarj-python-lint is 0.37.0
drift  pyproject.toml: sarj-python-lint==0.25.0  --  installed sarj-python-lint is 0.37.0
ok     pyproject.toml: sarj-lint-configs==0.32.0  --  matches the installed wheel
drift  package.json: @sarj/eslint-plugin@2.16.0  --  the bundled eslint.strict.mjs is tested against 9.0.0

It reports; it never rewrites. Exit 1 on drift.

The sibling linter versions are not yours to pick. sarj-lint-configs pins sarj-python-lint, sarj-sql-lint and sarj-iac-lint exactly, so doctor reads them out of the wheel you already installed and derives what every other site should say — including the pre-commit tag, which lives in a different namespace (python-v0.37.0 for sarj-lint-configs 0.32.0) that nobody should have to translate by hand.

The block init writes has no rev: at all. One repo: local hook runs the CLI from the environment your pyproject.toml pin already fixed, which deletes the second pin site and starts the toolchain only once per commit:

repos:
  - repo: local
    hooks:
      - id: sarj-standards-check
        name: sarj standards -- staged checks
        entry: uv run --frozen sarj-standards check --staged
        language: system
        verbose: true
        files: '(?i)(\.py|\.tsx?|\.jsx?|\.sql|\.tf|\.tfvars|\.hcl|\.ya?ml|\.toml|\.jsonc|\.mdx?|\.(?:bash|cfg|conf|env|ini|properties|sh|tftpl|zsh)|(?:^|/)\.env(?:\..*)?$|(?:^|/)(?:Dockerfile(?:\..*)?|Gnumakefile|Justfile|Makefile))$'

That is the block a Python repo gets. A TypeScript-only repo gets the same hook with uvx --from sarj-lint-configs==<version> instead of uv run --frozen; check --staged routes only staged paths to their applicable linters in either ecosystem.

Removed and renamed rules — the upgrade that crashes

Deleting a rule is not a lint-level change for the repo that uses it. A flat config still naming it makes ESLint exit 2 before it reads a single file:

TypeError: Key "rules": Key "@sarj/prefer-setup-file-mocks": Could not find
"prefer-setup-file-mocks" in plugin "@sarj".

The whole repo stops linting. A pre-commit hook id that no longer exists fails the same way, and because the strict config sets reportUnusedDisableDirectives: "error", every orphaned // eslint-disable-next-line @sarj/<removed-rule> is an error of its own.

rule-ledger.json ships inside the wheel and records every rule identifier this toolchain has ever shipped, along with what became of it. doctor reads it and names every stale reference — in configs, in suppression baselines, in pre-commit, and in ordinary source — before the upgrade that would break them:

drift  eslint.config.mjs: @sarj/prefer-setup-file-mocks x1  --  no longer exists -- Delete the config entry and suppressions; there is no replacement.
drift  src/legacy.ts: @sarj/no-implicit-attribute-access x1  --  no longer exists -- ...
drift  .pre-commit-config.yaml: no-implicit-attribute-access x1  --  no longer exists -- removed in sarj-python-lint 0.37.0 ...
drift  .sarj-python-baseline.json: SARJ083 x1  --  no longer exists -- removed in sarj-python-lint 0.37.0 ...
drift  eslint.config.mjs: @sarj/strict-test-assertions x1  --  renamed to @sarj/prefer-whole-object-assertion -- the old name no longer resolves: @sarj/eslint-plugin 9.0.0 deleted the deprecated aliases 7.0.0 shipped ...

There is nothing to fall back on in any of these cases. A deleted rule, a SARJnnn code that was renumbered and a rule renamed without an alias all fail identically at load time, and the failure names the identifier but never says it went deliberately or what replaced it. The ledger is where that is written down.

The ledger is generated by make sync-rule-ledger, which never deletes: a rule that leaves a registry is retired, and tests in both this package and the plugin fail until the ledger matches the live registries again. So the next removal is recorded whether or not its author thought about consumers.

update --configs-only --check — synced configs are unmodified

Ruff cannot extend a config out of an installed package's path portably, which is why config refresh writes a copy into your repo instead of you referencing one. A copy can be edited, so update --configs-only --check compares each one byte-for-byte and fails CI when it differs. One consumer's copy of the ESLint config had quietly drifted to 120 rules against a canonical 145 — missing 30, carrying 5 that no longer exist upstream — and nothing caught it.

update --configs-only and its --check mode operate on the config set in .sarj-standards.toml, so a Python repo is not asked to carry an ESLint config it never wanted. That matters: sync --check used to insist on all six files, report permanent drift on the two a repo had no use for, and so never made it into anyone's CI.

Your own settings are not clobbered, and you do not need to fork anything.

• Ruff. [tool.ruff.lint] ignore in your pyproject.toml is additive over the extended file's list — verified, not assumed. Adding ignore = ["ANN001"] silences ANN001 and leaves all 43 canonical ignores in force. Put every local ruff decision in pyproject.toml and leave .ruff-strict.toml alone.
• ESLint. Your eslint.config.mjs spreads the synced array and appends. Flat config is last-wins, so an override block after ...strict relaxes a rule, scopes one to a directory, or adds a framework exemption — and still receives every rule added upstream. init writes that block for you, commented, with a unicorn/filename-case example, because "the canonical config does not know about my framework's filenames" is the most common reason people forked it.

ESLint peers

eslint.strict.mjs imports eleven npm packages, and the set does not resolve on its own. The unicorn floor (72) pulls eslint >= 10.4, while eslint-plugin-react@7.37.5 — the newest published release — peers eslint <= ^9.7. npm install exits ERESOLVE, so the config is unreachable until you add an overrides entry:

{ "overrides": { "eslint-plugin-react": { "eslint": "$eslint" } } }

Following the README used to mean hitting Cannot find package nine times and then that dead end, with nothing naming the escape. That is most of why TypeScript repos vendored the file. init writes the block; peers prints it.

uv run --frozen sarj-lint-configs peers   # prints the set and one install command

The set is pinned in eslint.peers.json inside the wheel, packages/typescript installs exactly it, and tests there both load the shipped config through a real ESLint and lint real files with it — so "these versions resolve and this config works" is a CI claim, not a sentence in a README.

Three of the pins are load-bearing floors rather than just "current":

eslint-plugin-zod returns to the config after being dropped in #155, when the one rule taken from it (zod/prefer-enum-over-literal-union) was replaced by @sarj/prefer-zod-enum and the import went with it. Two of its rules are worth the dependency on their own measured evidence — zod/prefer-nullish (691 hits across 12 of 17 audited repos, autofixable) and zod/no-any-schema (159 hits across 10) — and enabling a maintained upstream rule beats keeping a local copy of it. The rest of the plugin stays off: most of it is Zod v3 → v4 migration advice (no-number-schema-with-int, prefer-top-level-string-formats, the no-schema-with-is-* deprecations) that would misfire on a v3 consumer.

All 18 eslint-plugin-react rules remain active on ESLint 10. Version 7.37.5 calls rule-context APIs removed by ESLint 10, so the config wraps it with ESLint's official @eslint/compat adapter. Runtime tests lint real TSX and fail if the adapter or any configured React rule becomes inert.

You do not have to install Python to get the ESLint config. If your repo is TypeScript-only, either run init once from uvx and commit the result, or skip sarj-lint-configs and use the plugin's own preset:

// eslint.config.mjs
import sarj from "@sarj/eslint-plugin";
export default [sarj.configs.strict];

That preset carries the @sarj rules only — not the typescript-eslint, unicorn, react or perfectionist layers the shipped config adds — but it needs one npm package and no Python. It is flat-config shaped as of @sarj/eslint-plugin 6.0.0; before that both presets declared plugins in eslintrc array form and ESLint threw on sight, which is the other reason people copied the file.

Rules removed in a breaking release

An eslint-disable naming a rule that no longer exists reports "Definition for rule was not found" until the comment is dropped. 3.0.0 removed no-unsafe-cast, prefer-shadcn, no-sequential-await, require-schema-validate-search and single-public-export; 5.1.0 removed ban-loose-type-guards-in-tests, no-implicit-attribute-access and prefer-setup-file-mocks.

What each config lands as

Polyglot repos can route the two ecosystems to their own roots:

uv run --frozen sarj-standards update --configs-only --python-dest python --typescript-dest web --force

check — the custom rules

check discovers every rule from the exact registry versions installed with this package. Files and recursively discovered directory contents are routed to the applicable registry by suffix. It is deliberately zero-tolerance: it does not accept suppression baselines that a change could inflate to conceal new findings.

Do not copy the runner into consumer repositories. Keeping it inside the wheel ensures the CLI implementation and its exact registry dependencies upgrade as one tested unit.

Legacy config-path exports remain importable for compatibility. New automation should use sarj-standards show config NAME or the typed adoption libraries.

0.8.0 — PLC2701 moved out of ruff

PLC2701 import-private-name is in the ignore list. It cannot tell a private name of ours from a private name of a dependency's: its exemption is "same top-level package", so from livekit.agents.inference_runner import _InferenceRunner — an API livekit made private in 1.6.6, with no public replacement — is flagged identically to a first-party helper someone forgot to export. Ruff has no configuration surface that separates them.

The check is replaced by SARJ048 in sarj-python-lint, which resolves the imported module against your project tree and fires only on first-party modules. It runs as part of sarj-lint-configs check.

If you are not running sarj-lint-configs check, re-enable PLC2701 in your own [tool.ruff.lint] extend-select — an over-firing check beats no check.

Attribute access (session._stt) is unchanged: ruff's SLF001 and pyright's reportPrivateUsage both still fire, and neither can make the first/third-party distinction. Both configs carry the rationale and the escape hatches inline.