surface-recon

Map what a target exposes, what it demands of you, and what will bite you. Produce a report someone else can build from without repeating your work.

This skill investigates. It does not design a command surface or write code: that is cli-build's job. Stopping at a report is the point: recon tells you whether building is worth it.

The one rule

Every claim in the report is either observed or labeled as unverified. A recon report that guesses without saying so is worse than no report, because the implementer trusts it and loses a day. Mark inference as inference.

Concretely: the endpoint table holds requests you observed. Everything else goes under "Needs verification", each with the step that would confirm it.

Open recon/friction.md before Phase 0 and append to it as you go. Every time a playbook is thin, a command fails for a reason its help text did not predict, or a gate does or does not hold, that is one line written at the moment it happens. It ships with the report and it is the only input that improves this skill. See friction-log.md.

The tool

Observation happens through agent-browser, and that tool ships its own guides that stay version-matched with the binary. Load them rather than guessing commands from memory: