HookGuard – scanner for malicious Claude.md and agent config files
Catches invisible Unicode tricks and RCE hooks in CLAUDE.md files.
The missing linter and lsp for AI coding assistants. Validate CLAUDE.md, AGENTS.md, SKILL.md, hooks, MCP. Plugin for all major IDEs included, with autofixes.
Makes agent configs first-class with 229 domain-specific rules, autofix, and LSP support — so a tiny syntax mistake stops being a silent failure. The cross-editor plugins and GitHub Action are the standout moves: lint in your IDE and enforce checks in CI. I want a clearer map of which rules target which toolchains, but the breadth of integrations is impressive.
Developers and toolsmiths who build, deploy, or maintain AI agents/skills (agent engineers, extension authors, CI maintainers)
Catches invisible Unicode tricks and RCE hooks in CLAUDE.md files.
Hook guard stops Claude from bypassing commit rules when confirming changes.
Syncs MCP configs across Cursor, Copilot, Claude—but still fragmented tool ecosystem.
Four focused API calls generate better configs than single-prompt approaches.
Profile manager for AI agents — switch configs like dotfiles but for Claude and Cursor.
Auto-memory hooks for Claude Code, but context window bloat within sessions remains unsolved.