首先开源库都会实现某个功能，我们先要理解这个库是干嘛的
Find AILANG vs Python eval gaps and improve prompts/language. Use when user says 'find eval gaps', 'analyze benchmark failures', 'close Python-AILANG gap', or after running evals.
Monitor and triage GitHub issues against design docs and implementation status. Use when user asks to "check issues", "triage issues", "sync issues", "what issues are open", or wants to ensure issues are up-to-date with development progress.
Analyze design docs, calculate velocity from recent work, and create realistic sprint plans with day-by-day breakdowns. Use when user asks to "plan sprint", "create sprint plan", or wants to estimate development timeline.
Guides development of AILANG builtin functions. Use when user wants to add a builtin function, register new builtins, or understand the builtin system. Reduces development time from 7.5h to 2.5h (-67%).
Expert in Google Merchant API (the new API replacing Content API). Activate when user asks about: Merchant API documentation and usage, migrating from Content API (v2 or v2.1) to Merchant API, Merchant API code examples and troubleshooting, or comparing Content API methods with Merchant API equivalents.
让PPT更有说服力
Add or modify tasks in Implementation Plan. Use when user proposes new requirements, changes priorities, or adjusts existing task scope.
This skill should be used when tracing ManifestWork resources through the Maestro system to find relationships between user-created work names, resource IDs, and applied manifests, or to debug manifest application issues across the management cluster and database.
Help address review/issue comments on the open GitHub PR for the current branch using gh CLI; verify gh auth first and prompt the user to authenticate if not logged in.
Practical guide for building safe, syntax-aware srgn CLI commands for source-code search and transformation. Use when users ask for srgn commands, scoped refactors (comments/docstrings/imports/functions), multi-file rewrites with --glob, custom tree-sitter query usage, or CI-style checks with --fail-any/--fail-none.
Effective code search, analysis, and refactoring using ast-grep (sg). Use this skill for precise AST-based code modifications, structural search, and linting.
Inspect GitHub PR checks with gh, pull failing GitHub Actions logs, summarize failure context, then create a fix plan and implement after user approval. Use when a user asks to debug or fix failing PR CI/CD checks on GitHub Actions and wants a plan + code changes; for external checks (e.g., Buildkite), only report the details URL and mark them out of scope.
Create new publish/subscribe examples for ROS2 message types. Use when the user wants to add a new example, create pub/sub for a new message type, or add trajectory/sensor/geometry message examples.
Map permissible values in LinkML enums to ontology terms. Use this skill when: (1) Adding or updating ontology mappings (meaning: field) for enum permissible values, (2) Fixing validation errors from linkml-term-validator, (3) User asks to map enums to ontology terms or fix CURIE mappings. This skill covers OAK/runoak lookup, CURIE verification, and validation workflows.
Claude Code Skills and Agent Architecture in DiPeO
基于当天的 Intake 文件生成可执行计划 `plan_<topic>.md`，并落盘到 `<MM_DD>/codex/plan/`；计划包含 Tasks+DoD+Dependencies、可选 Gates、Deliverables（当天目录路径）与 Rollback。适用于“基于intake写计划/生成plan_<topic>.md/补DoD和门禁”等请求。
将一次执行的改动/命令/验证/产物整理成结构化 `manifest_<topic>.json`，并落盘到 `<MM_DD>/codex/artifact/`，用于证据链与可追溯交付。适用于“写manifest/记录证据链/把今天做了什么落成json/生成artifact清单”等请求。
批量推进编排 Skill：把“一堆事项/bug/想法/审稿意见”收口为 Assumptions → Triage Queue → Sprint Plan（含并行组/DoD/Gates/回滚）→ Next Backlog；可选将结果落盘到 `<MM_DD>/codex/intake/intake_<topic>.md`、`<MM_DD>/codex/plan/plan_<topic>.md`（以及 `<MM_DD>/codex/exec/exec_<topic>.md`）。适用于“我有一堆事想批量推进/做个冲刺计划/把 backlog 变成可执行任务”等请求。
Expert in quantifying damages from construction and proximity impacts including noise, dust, vibration, traffic disruption, visual impairment, and business losses. Use when assessing temporary construction impacts, permanent proximity damages, business loss claims, or quantifying injurious affection beyond general principles. Key terms include injurious affection, construction impacts, noise modeling, vibration damage, traffic disruption, visual impairment, business losses