菜单
Improve test coverage in the OpenAI Agents Python repository: run `bun run coverage`, inspect coverage artifacts, identify low-coverage files, propose high-impact tests, and confirm with the user before writing tests.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Use when the user asks for Codex Security workflows: repository-wide or scoped-path security scans, security reviews of pull requests, commits, branch diffs, or working-tree patches, deep exhaustive repository scans, fixing validated or plausible security findings, or individual threat-modeling, finding-discovery, validation, and attack-path-analysis phases.
Clone and inspect external repositories in a reusable local exploration cache. Use this skill when the user asks to explore, inspect, investigate, compare, or answer questions about a repository that may not already be in the current workspace.
Create atomic git commits
Use when diagnosing or fixing runtime CPU spikes, high event-loop latency, memory growth, heap leaks, garbage-collection pressure, or OOM crashes in Node.js, Bun, Deno, JavaScript/TypeScript services, or comparable long-running runtimes. Includes code-first triage, guided CPU and memory profiling, subagent-orchestrated review, before/after profile comparison, and structured remediation reports.
SQLite database optimization, query writing, indexing, and best practices specialist. Proactively analyzes and optimizes SQLite databases for performance and reliability.
Run an extremely strict maintainability review for abstraction quality, giant files, and spaghetti-condition growth. Use for a thermo-nuclear code quality review, thermonuclear review, deep code quality audit, or especially harsh maintainability review.
| name | test-coverage-improver |
| description | Improve test coverage in the OpenAI Agents Python repository: run `bun run coverage`, inspect coverage artifacts, identify low-coverage files, propose high-impact tests, and confirm with the user before writing tests. |
Use this skill whenever coverage needs assessment or improvement (coverage regressions, failing thresholds, or user requests for stronger tests). It runs the coverage suite, analyzes results, highlights the biggest gaps, and prepares test additions while confirming with the user before changing code.
bun run coverage to regenerate coverage/ data.coverage/ .tests/, rerun bun run coverage, and then run bun run check before marking work complete.bun run coverage at repo root. Avoid watch flags and keep prior coverage artifacts only if comparing trends.bun run coverage for file-level totalssrc/ before examples or docs.tests/ and avoid flaky async timing.scripts/, references/, or assets/ unless needed later.bun run coverage instead of guessing.