以考官视角做“任务是否完成”审计:列出验收标准→找不确定点→补测试/证据→修复→重复直到无剩余疑虑。用户要求 100% 确信或准备交付/提 PR 时调用。
How this skill is triggered — by the user, by Claude, or both
Slash command
/polyglot-agent-harness:completion-examinerThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
- 用户要求“你是否 100% 确信实现正确/是否完成”
把“我觉得差不多”变成“有可复现证据链的完成结论”:
列验收标准
做不确定性清单
把疑点变成可验证项
验证并收敛证据
复盘剩余风险
交付门禁(宣称“完成”前必过)
pr-workflow 的 blocking order 逐项通过:cargo fmt --all -- --check → clippy(warnings denied)→ 所改 crate 的单元/集成测试 → E2E(含可复制命令与期望输出)pr-workflow(PR 草稿、手动提交策略)与 report-generator(本地报告更新);子 agent 在最终总结开头须声明所应用的 skill(如 Applying pr-workflow),见 pr-workflow step 0cargo fmt,除非用户明确要求)auto-skill-maintainer 更新技能并提交Provides behavioral guidelines to reduce common LLM coding mistakes, focusing on simplicity, surgical changes, assumption surfacing, and verifiable success criteria.
Searches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.
Creates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.
npx claudepluginhub jiaxin2006/polyglot-agent-harness --plugin polyglot-agent-harness