gregce
UserRun your coding agent unattended, and trust the result. A supervision harness for the the agent you already use. Isolated sandbox, durable per-turn state, and a watchdog gate the agent can't fool. Output is an auditable artifact.
Categories
Indexed Skills (7)
live-narrator
Amends and extends the live run narrative one beat at a time while a deadreckon run is in flight. Output is one JSON object.
narrator-as-built
Produces RUN-AS-BUILT.md for a deadreckon run using stoa subsystem-AS-BUILT conventions.
narrator-decisions
Produces RUN-DECISIONS.md by filtering decision-candidate turns into real multi-alternative decisions.
narrator-overview
Produces overview sections for deadreckon RUN-NARRATIVE.md in the stoa implementation-doc shape.
run-narrator
Produces RUN-NARRATIVE.md, RUN-AS-BUILT.md, RUN-DECISIONS.md, and optional AS-BUILT-DELTA.md from a deadreckon run trace and diff. Output is one JSON object.
narrator-phases
Produces phase-by-phase prose for deadreckon RUN-NARRATIVE.md from turn records and diff samples.
default-coding
Default unattended coding skill for deadreckon V0.
Bio shown is the top-scored skill's repo description as a fallback — real GitHub bios land in a future update.