Reference · Sheet 02

CLI

Commands

CommandWhat it does
hyperfixer run [flags]Run gates in cost order, write the verdict, exit by the contract below
hyperfixer fix [flags]Run every gate's fixCommand (biome --write, eslint --fix), then verify with the normal pipeline
hyperfixer initDetect the stack (Biome or ESLint, tsconfig, Bun or vitest or npm test) and write a tailored hyperfixer.config.json
hyperfixer hintPrint the first actionable fix from the last verdict, warns on stderr when the verdict is older than 10 minutes
hyperfixer doctorCheck toolchain and config health
hyperfixer install-hooksInstall git pre-commit (cost <= 50) and pre-push (full pipeline) hooks, never clobbers foreign hooks
hyperfixer install-claudeInstall the Claude Code PreToolUse hook into .claude/settings.json
hyperfixer claude-hookInternal, PreToolUse entry point reading the hook payload from stdin

Run flags

FlagEffect
--jsonMachine-readable verdict on stdout
--quietNo human output, verdict.json still written
--config <path>Config file, default hyperfixer.config.json
--only <a,b>Run only the named gates
--max-cost <n>Run only gates with cost <= n
--changedExpand {changed} in commands to git-changed files
--no-cacheIgnore and do not update the input-hash cache
--no-fail-fastRun all gates even after a failure
--out-dir <dir>Verdict output directory, default .hyperfixer

Exit codes

CodeMeaning
0All gates pass
1A gate failed, the code is wrong, read the hint
2Setup problem: config error, empty pipeline (zero gates executed), or another run holds the lock
3Gate infrastructure failed (timeout, tool crash), do not edit code

The contract is strict by design: a broken config is exit 2 and a hung tool is exit 3, never exit 1, so an agent reading exit codes never mistakes a setup or infrastructure problem for a code problem.

Robustness