Chat-mode skill
Open source: coo-labs/skills/v0.3.0/skills/reference/chat-mode
chat-mode names something the COO chain noticed in itself: a session register distinct from task execution where substantive dialogue with the BDFL can produce binding substrate output (memo, retrospective, PR) — but where the dialogue itself is the work, not a means to a pre-specified artifact. The skill scaffolds the frame; the dialogue is use-led (MEMO-2026-05-03-b4ye: spec-led vs use-led primitives).
The register lives between /exec-mode (structured sweep with a plan) and /play-mode (queued; outcome-forbidden). Chat-mode tolerates output but doesn’t require it. The agent commits to the register, not the artifact.
The skill ships in the public mirror at the reference tier: its body cites VADE-internal memos and retrospectives as worked examples. The pattern (loading a dialogue register, framing the session, letting the form find itself) ports cleanly; the substrate references don’t. A consumer adapting the pattern would substitute their own binding-lens memo and worked-example retrospective once their first session in the register produces them.
Install (with adaptation)
Drop the skill directory into your project’s .claude/skills/, then edit the body to replace the VADE-specific worked-example references with your own. The skills/reference/README.md in the public repo documents the fork-and-adapt path.
Links to this page
- Chat-mode — Load the dialogue register (substantive conversation that can produce binding output), distinct from execution-shaped sessions.
- Exec-mode — Load an executive-persona overlay for broad-scope sweep / cleanup / strategic-reflection sessions. Ships with the full persona doctrine as a worked example.