Hook Claude Code
Runs when you submit a prompt, before the agent sees it, executing ledger-capture.py, inbox-drain.py, telegram-reply-directive.py and provenance-gate.py via python3 (4 commands). From Szotasz/marveen.
Hook Claude Code
Runs when you submit a prompt, before the agent sees it, executing ledger-capture.py, inbox-drain.py, telegram-reply-directive.py and provenance-gate.py via python3 (4 commands). From Szotasz/marveen.
Hook Claude Code
Runs after a tool call finishes for mcp__plugin_telegram_telegram__reply tool calls, executing ledger-outbound.py and tool-log-capture.py via python3 (2 commands). From Szotasz/marveen.
Hook Claude Code
Runs before the agent uses a tool for Bash, .*send_email.* and mcp__plugin_telegram_telegram__reply tool calls, executing outgoing-copy-gate.py via python3 (3 commands). From Szotasz/marveen.
Hook Claude Code
Runs when the agent finishes a response, executing telegram-reply-guard.py via python3. From Szotasz/marveen.
Hook Claude Code
Runs when a session starts on startup, resume and clear, executing ledger-replay.py via python3. From Szotasz/marveen.
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: