Plugin Claude Code
The watchdog for scheduled Claude Code agents: proof the run did the job, and an alert the moment it doesn't — missed runs, silent failures (green != done), retry storms, runaway cost, drift. Email/Telegram/Slack/webhook.
Agent
Reviews the health of the user's scheduled agents via RunVouch and explains what needs attention and why. Use when the user asks "are my agents ok", "what failed last night", or before a morning run that depends on nightly jobs.
Command
Show RunVouch status of all scheduled agents and open alerts.
Hook
Runs when a session starts, executing rv-hook.sh. From runvouch/runvouch.
Hook
Runs after a tool call finishes, executing rv-hook.sh. From runvouch/runvouch.
Hook
Runs when the agent finishes a response, executing rv-hook.sh. From runvouch/runvouch.
Skill Claude CodeCodex
Check RunVouch before trusting the output of another scheduled agent, register/wrap jobs with rv, and interpret RunVouch alerts (MISSED, FAILED, NOEVIDENCE, RETRYSTORM, BUDGET, DRIFT, STALLED).
Skill Claude CodeCodex
Wrap a scheduled OpenClaw task in a RunVouch run. Reports start and end, attaches the output file as evidence, and alerts on Telegram, Slack or e-mail when the task is missed, fails, or ends without the file it was supposed to write.
MCP server Claude CodeCodexCursor
Watchdog for unattended AI agents: alerts, evidence checks and a verifiable proof per run. Remote server at api.runvouch.com.