registry / ratchet-cc / 0.6.4

ratchet-cc@0.6.4

Every correction you give your coding agent becomes an enforced check it can never violate again. CLAUDE.md is advice; Ratchet is law.

AI Security Review

scanned 2h ago · by lpm-firewall-ai

LPM treats this as warn-only first-party agent extension lifecycle risk. No install-time attack behavior is present. Explicit `ratchet install` configures project-local AI-agent lifecycle hooks that invoke this package, creating a guarded extension lifecycle risk.

Static reason

No blocking static signals were detected.

Trigger

User runs `ratchet install` or explicitly installs the Claude plugin/hooks.

Impact

The package code runs for configured agent tool, prompt, and stop events and can inspect local project rules/session inputs.

Mechanism

Project-local Claude/Cursor/Codex hook registration and optional pre-commit hook installation.

Rationale

Source inspection finds no malicious install-time or stealth behavior, but the package deliberately installs project-level agent lifecycle hooks when requested. Per policy, this first-party agent-extension setup warrants a warning rather than a block.

Evidence

package.jsonbin/ratchetsrc/cli.jssrc/hooks/install.jssrc/hooks/runtime.jssrc/remote-pack.jssrc/selfupdate.jssrc/mine.js.claude/settings.json.cursor/hooks.json.codex/config.toml.git/hooks/pre-commit.ratchet/rules/*.yaml

Network endpoints2

/repos/${owner}/${repo}/contents/${path}

ratchet-cc@0.6.4

AI Security Review

Decision evidence

Source & flagged code

Source & flagged code

Findings

Source & flagged code