Catch more with static checks: type-aware oxlint, oxfmt, knip

.github/workflows/ci.yml

@@ -5,22 +5,33 @@ on:
     branches: [main]
   pull_request:
 
+concurrency:
+  group: ci-${{ github.ref }}
+  cancel-in-progress: true
+
 jobs:
-  check:
+  # Fast, static checks first; most mistakes stop here before any test runs.
+  static:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v5
+      - uses: ./.github/actions/setup
+      - run: bun run format:check
+      - run: bun run lint
+      - run: bun run typecheck
+      - run: bun run knip
+      - run: bun run build
+      - name: Run the bundled CLI under Node
+        run: node dist/index.mjs --version
+
+  test:
+    needs: static
     strategy:
       fail-fast: false
       matrix:
         os: [ubuntu-latest, macos-latest]
     runs-on: ${{ matrix.os }}
     steps:
       - uses: actions/checkout@v5
-
       - uses: ./.github/actions/setup
-
-      - run: bun run lint
-      - run: bun run typecheck
       - run: bun run test
-      - run: bun run build
-
-      - name: Run the bundled CLI under Node
-        run: node dist/index.mjs --version

.oxfmtrc.json

@@ -0,0 +1,5 @@
+{
+  "$schema": "./node_modules/oxfmt/configuration_schema.json",
+  "printWidth": 100,
+  "ignorePatterns": ["bun.lock"]
+}

.oxlintrc.json

@@ -0,0 +1,47 @@
+{
+  "$schema": "./node_modules/oxlint/configuration_schema.json",
+  "plugins": ["eslint", "typescript", "unicorn", "oxc", "import", "node", "vitest"],
+  "categories": {
+    "correctness": "error",
+    "suspicious": "error",
+    "perf": "error"
+  },
+  "rules": {
+    "typescript/no-floating-promises": "error",
+    "typescript/no-misused-promises": "error",
+    "typescript/await-thenable": "error",
+    "typescript/require-await": "error",
+    "typescript/switch-exhaustiveness-check": "error",
+    "typescript/no-unnecessary-condition": [
+      "error",
+      {
+        "allowConstantLoopConditions": "only-allowed-literals"
+      }
+    ],
+    "typescript/no-unnecessary-type-assertion": "error",
+    "typescript/restrict-template-expressions": "error",
+    "typescript/prefer-nullish-coalescing": [
+      "error",
+      {
+        "ignorePrimitives": {
+          "string": true
+        }
+      }
+    ],
+    "typescript/no-explicit-any": "error",
+    "typescript/no-non-null-assertion": "error",
+    "typescript/consistent-type-imports": "error",
+    "import/no-commonjs": "error",
+    "import/no-cycle": "error",
+    "eslint/no-console": "off",
+    "eslint/no-await-in-loop": "off",
+    "oxc/no-map-spread": "off",
+    "vitest/valid-expect": [
+      "error",
+      {
+        "maxArgs": 2
+      }
+    ]
+  },
+  "ignorePatterns": ["dist", ".scratch"]
+}

AGENTS.md

@@ -6,7 +6,7 @@ syncode is a TypeScript CLI, published to npm, that links AI coding agents' inst
 
 - Package manager: bun. Runtime: Node >= 22.12 for the published CLI, >= 22.18 to run from source.
 - `bun run dev -- <command>`: run from source with Node's type stripping.
-- `bun run check`: lint (Biome), typecheck (`tsc`), test (Vitest), build (tsdown). Run before pushing.
+- `bun run check`: format check (oxfmt), type-aware lint (oxlint), typecheck (`tsc`), unused code (knip), tests (Vitest), build (tsdown). Run before pushing; `bun run format` fixes formatting.
 
 ## Layout
 
@@ -21,6 +21,7 @@ syncode is a TypeScript CLI, published to npm, that links AI coding agents' inst
 
 - Only Node built-ins plus `@clack/prompts`. Prefer data over per-agent classes.
 - Never prompt when stdin or stdout isn't a TTY; keep the local file and report it instead.
-- Expected failures throw `UserError` (printed without a stack). No `require`: the bundle is ESM.
-- Test through the CLI boundary in `tests/`. Don't export internals or add flags just for tests.
+- Expected failures throw `UserError` (printed without a stack). No `require`: the bundle is ESM (enforced by `import/no-commonjs`).
+- Prefer static checks over tests: if a type, a lint rule or knip can catch a mistake, use that. Tests are for behavior only the real CLI can show (filesystem effects, non-interactive fallbacks, migration).
+- Test through the CLI boundary in `tests/`, asserting on filesystem state and `status --json`, not log text. Don't export internals or add flags just for tests.
 - Comments explain why, not what.

CONTRIBUTING.md

@@ -14,7 +14,8 @@ bun run dev -- --help
 ## Before opening a pull request
 
 ```bash
-bun run check   # lint, typecheck, test, build
+bun run format  # fix formatting
+bun run check   # format check, lint, typecheck, knip, test, build
 ```
 
 [AGENTS.md](AGENTS.md) describes the layout and conventions. To support a new agent, add an entry to `src/agents.ts` and the README table, with a link to the docs showing where it reads instructions and skills.

README.md

@@ -4,6 +4,7 @@
 [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](LICENSE)
 
 Keep your AI coding agents' instructions, skills and config files in one git repo, and link them into every agent on every machine.
+
 - **One set of instructions** for Claude Code (`CLAUDE.md`), Codex and OpenCode (`AGENTS.md`) and Gemini CLI (`GEMINI.md`), with optional per-OS and per-machine parts.
 - **One skills folder** linked into `~/.agents/skills` (Codex, OpenCode, Gemini CLI, Cursor, Copilot CLI) and `~/.claude/skills`.
 - **Only what you choose.** Config files are synced only after you `adopt` them, and credentials like `auth.json` are refused, so sessions, caches and tokens never reach git.
@@ -75,6 +76,7 @@ If a local file differs from the repo copy, `apply` asks whether to keep it, rep
 ## Upgrading from 1.x
 
 Run `syncode init`. It detects the 1.x config and:
+
 1. Replaces the 1.x whole-folder symlinks (such as `~/.config/opencode`) with real copies, so every agent keeps working.
 2. Moves `.agents/skills/*` to `skills/`, and takes the first `AGENTS.md` or `CLAUDE.md` it finds in `configs/` as `instructions/AGENTS.md`.
 3. Links everything the 2.0 way.
@@ -86,7 +88,7 @@ The 1.x `configs/` folder stays in the repo. `syncode adopt` the files you want
 ```bash
 bun install
 bun run dev -- status   # run from source (Node >= 22.18)
-bun run check           # lint, typecheck, test, build
+bun run check           # format, lint, typecheck, knip, test, build
 ```
 
 See [AGENTS.md](AGENTS.md) for how the code is organized.