AI Agent Skills

Splits a batch of expensive per-item agent work into a cheap reversible pass over every item and an expensive irreversible pass, separated by a method-freeze checkpoint.…

Map the buying committee at one target account - economic buyer, champion candidates, technical evaluators, users and blockers (legal, security, procurement, finance) -…

Read-only data quality audit of the connected CRM - field completeness per mapped property, duplicate candidates, stale records, enrichability estimate. Produces a repor…

Use when writing or reviewing any API endpoint that retrieves, updates, or deletes a record by user-supplied ID. Triggers on route handlers with :id params, controller f…

Dialogue-driven LinkedIn positioning sparring partner. Goal-first — audits WHICH specific professional outcome this profile serves BEFORE drafting. Same person produces…

Whole-repo audit for over-engineering. Like ponytail-review, but scans the entire codebase instead of a diff: a ranked list of what to delete, simplify, or replace with…

Validate features before training - IC significance, stability, redundancy, and contamination checks. Use when auditing feature quality before model fitting.

依赖供应链审计：pnpm/npm audit 输出与退出码解读、license 与投毒风险检查清单、锁文件漂移检测命令。任务要求审计项目依赖的已知漏洞、许可证风险、可疑包或 lockfile 一致性并出结论时用；单独安装/升级某个依赖或纯功能开发不展开本流程。

agent 环境安全事件响应：分类→控制蔓延→取证留痕→恢复→复盘的分步处置流程，覆盖密钥泄露、提示注入触发、依赖投毒、未授权操作四类事件，每条处置附命令证据。DSH/agent 环境出现疑似安全事件需要按流程处置与复盘时用；日常开发与例行维护不用。

仓库/软件安全审计总览：范围界定→资产清单→风险分级→逐项验证→报告模板的分阶段流程，按需转调 secret-scan、dependency-audit、supply-chain-review、prompt-injection-review 四个专项技能。用户要求整体审计仓库、规划审计步骤或汇总多类发现出报告时用；只查密钥/依赖等单一主…

新功能/新系统的轻量威胁建模：固定对象→划定范围与信任边界→资产清单→STRIDE 逐资产威胁表→攻击树（可选）→缓解与优先级，产出可进设计评审的威胁模型文档。为新功能/新改动做设计阶段安全评审、梳理信任边界或画攻击树时用；与安全边界无关的纯 bug 修复、已有成熟建模流程的团队不展开本流程。

Use after a discrete event, milestone, or push has completed — successful or not. Runs a structured after-action review — what was expected, what happened, why they diff…

Enforce clean-text and no-hidden-marker requirements, audit inspectable characters and provenance, and state the verification boundary for statistical text marks. Use wh…

Refine an existing prompt artifact — inline prompt text, markdown prompt files (system prompts, agent instructions, slash commands), or a complete skill directory — agai…

Guides building evaluation harnesses, regression test suites, and runtime guardrails for LLM agents. Use when a user asks to "evaluate this agent," "write test cases for…

Google Workspace Admin SDK: Audit logs and usage reports.