Repository Audit with Claude Code: API Runtime Pattern
A production playbook for repository audit in cross-industry operations using Claude Code: api runtime pattern, run-scoped inputs, logs, typed results, and artifacts.
Audience: Engineering platform teams
The problem
Engineering platform teams need repository audit to run repeatedly against source trees, diffs, dependency manifests, and test output. In cross-industry operations, the pain is not one good answer; it is repeatability, auditability, exception handling, and evidence that survives handoff.
Implementation path
Package the repository audit instructions as a skill, send source trees, diffs, dependency manifests, and test output as run-scoped inputs, execute with Claude Code, poll terminal status, and consume argo.result.v1 instead of parsing a transcript.
Tradeoffs and failure modes
The API boundary forces the workflow to define inputs, terminal states, and result shape before customers depend on it. For repository audit, the practical test is whether a second run can be debugged, retried, and consumed by a product without reading the raw agent transcript.
Run request
POST /api/skills/<skill_id>/run
provider=claude-code
workflow=repo-audit
inputs[]=@./input-pack.zip
result_schema=argo.result.v1
Run this on Argo