executing-plans

Use when CODEX selects ECI or active ATE routes bounded work through ECI

Use when writing or verifying project-understanding ledgers, context ledgers, ECI/ATE session ledgers, handoff context, or stop-hook ledger updates — keeps…

Use when working with Android phones or tablets — fastboot, adb, flashing, kernel updates, device debugging, on-device experiments, benchmarks, perf…

Use when debugging and needing falsifiable hypotheses, alternative explanations, or stronger root-cause discipline

Use when writing, reviewing, or modifying Go code (*.go, go.mod, go.sum files) or reviewing diffs/PRs containing Go changes — ensures code quality and…

Use when porting code, features, or capabilities from one codebase/project to another — extracts exhaustive feature inventory, analyzes gaps against target…

Use only when the user explicitly asks to use ATE or agent-teams-execution; mentioning or maintaining this skill alone does not invoke it.

Use only when the user explicitly asks to interactively brainstorm before implementation

Use when work is blocked, a review loop cannot converge, an ECI or ATE work-progress flow may escalate to the user, or a stop-hook bypass is being considered…

Use when deadlines, demos, incidents, launches, or expiring opportunities make the durable solution risky to finish in time

Use when Codex CLI/session has just relaunched after a crash with unfinished work or active spawned agents.

Use when implementation is complete, all tests pass, and you need to decide how to integrate the work - guides completion of development work by presenting…