Directory / hookify
hookify
StrongCreates Claude Code hooks from natural language — analyze your conversation for past mistakes or describe a pattern you want to block, and hookify writes the rule.
Verdict
Hookify turns Claude into a meta-tool for managing Claude's own behavior: run /hookify to analyze the current conversation and surface patterns worth preventing, or describe a behavior directly and get a hook rule file in .claude/. The rule engine uses regex matching against tool names and input fields — no code execution, no network calls. Rules take effect on the next tool use without a restart. The /hookify-list, /hookify-disable, and /hookify-delete management commands are included. The clearest use case: you've corrected Claude three times this session; run /hookify and it reads the conversation and writes the prevention rule for you.
Install
Audit receipts
Every listing is verified against a specific commit. These are the receipts.
Python hook scripts (UserPromptSubmit, PreToolUse, PostToolUse, Stop) read .claude/hookify.*.local.md rule files and evaluate regex conditions. No shell=True, no eval(). hooks.json is the hook registry.
grep matched rule_engine.py for "eval" — false positive from method name _regex_match and re.compile(). Actual code uses Python re module only. Transcript path read from hook input is an OS open() call on the user's own transcript file — expected and documented.