Hook Claude Code
Runs before the agent uses a tool for Bash, Write, Edit, MultiEdit, Skill and ^mcp__linear__save_issue$ tool calls, executing claude-hooks-log-wrapper.sh, env-read-guard.mjs, dogfood-skill-telemetry.sh and linear-issue-creation-guard.mjs with --command, --validate-safety, --prepare-resources, --file…
Hook Claude Code
Runs after a tool call finishes for Bash, Write, Edit, MultiEdit, Task and Skill tool calls, executing claude-hooks-log-wrapper.sh and dogfood-skill-telemetry.sh with --command, --success, --track-metrics, --file and --format (4 commands). From smith-horn/skillsmith.
Hook Claude Code
Runs when a tool call fails for mcp__skillsmith tool calls, executing session-mcp-disconnect-guard.sh. From smith-horn/skillsmith.
Hook Claude Code
Runs before the context is compacted for manual and auto tool calls, running an inline shell check (2 commands). From smith-horn/skillsmith.
Hook Claude Code
Runs when a session starts on startup, executing session-start-priming.sh, session-start-audit.sh and session-start-mcp-command-guard.sh (3 commands). From smith-horn/skillsmith.
Hook Claude Code
Runs when the agent finishes a response, executing session-stop-hook-safe.sh. From smith-horn/skillsmith.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: