What problem does it solve? Onboarding to an unfamiliar codebase is slow and error-prone: READMEs are stale, and guessing at architecture leads to wrong assumptions. This Skill produces a trustworthy briefing — what the repo does, one diagram, its external dependencies, and an explicit list of what could not be determined — by tracing real call paths instead of guessing. ## Core Features & Use Cases - Call-path tracing via CodeGraph: Seeds from entrypoints found in manifests and config, then follows one real path end to end using the codegraph CLI, with a documented fallback to reading entrypoints when no index exists. - Structured five-section report: A plain-English summary, an architecture diagram (delegated to the archify skill or a mermaid fallback), a parts table with file:line citations, an external-services inventory with evidence, and a mandatory ❓ Unclear section. - Honesty-first rules: Read-only operation, file:line citations for structural claims, and anything inferred rather than read is listed as an open question instead of a fabricated answer. - Use Case: A developer joins a team and runs /explain-repo on a service they have never seen; they get a half-page briefing, a diagram, and a list of questions to ask the original authors. ## Quick Start Ask the AI to run /explain-repo on the current repository to get a plain-English architecture briefing with a diagram and a list of unknowns.
