agent-baton

agent-baton is an Agent runtime & coordination infrastructure skill for AI agents, published by oborchers in the fractional-cto repository on GitHub. Skillspace rates agent-baton F (45/100) and the safety scan rated it risky; the rating is based on static analysis only.

Risky. This skill has patterns associated with unsafe behaviour. Read the source carefully before running it.

This skill should be used when work must be chained across two independent agent processes that share no parent session — e.g. one agent in one terminal finishes a task and a second agent (possibly a different tool entirely: Claude Code, Codex, Cursor, Gemini) must start only once the first is done. Covers passing a baton (publishing a completion signal), attaching an optional free-form payload for the next agent to read as context, waiting for a baton (blocking until it appears, with a deadline), the signal file format, atomic publish/claim rules, payload integrity and untrusted-content handling, and the failure modes of file-based agent coordination. Triggers on: 'pass the baton', 'pass a payload to the next agent', 'wait for the other agent', 'signal when done', 'start after the other agent finishes', 'chain two agents', 'hand off results to another agent process', 'notify the other terminal'.

Grade

F

Quality

45 / 100

Safety

Risky

GitHub stars

25

Unproven: rated by static analysis only, not yet compared head-to-head.

This rating describes the version of the repository we crawled; it may have changed since.

Skillspace indexes and rates this skill; it does not host or endorse it. Run third-party skills at your own risk.