Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add cass-2003/local-workflow-skill --skill agent-briefinggit clone --depth 1 https://github.com/cass-2003/local-workflow-skillWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/cass-2003/local-workflow-skill/agent-briefing)<a href="https://agentmods.dev/skills/cass-2003/local-workflow-skill/agent-briefing"><img src="https://agentmods.dev/badge/skills/cass-2003/local-workflow-skill/agent-briefing.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00105 | $0.12234 |
| Opus 5 | $0.00053 | $0.06117 |
| Sonnet 5 | $0.00021 | $0.02447 |
| Haiku 4.5 | $0.00011 | $0.01223 |
Grade A, and why
agent-briefing scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 7d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 480 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent 提示词简报
Agent 提示词简报(agent-briefing,兼容 slug: apb)负责本技能描述范围内的定位、执行、验证和交接边界;旧短 slug 仅作兼容 alias/URL 主键,不作为规范技能名。
定位:把“让 Agent 去做事”收敛成可委派、可执行、可审计、可验收的任务简报,而不是润色一段提示词。 铁律:没有目标、上下文、边界、权限、停止条件、验收格式,不生成执行型 Agent brief;缺证据时输出缺口,不把假设包装成任务事实。
快速总则
- 先判任务类型:只读搜索、代码理解、方案规划、实现修改、测试验证、代码审计、外部调研、远程执行、长任务监控、多 Agent 协作分别生成不同 brief。
- Agent brief 的目标是降低跑偏:明确做什么、不做什么、看什么、不能假设什么、允许用什么工具、何时停止、如何回传。
- 子 Agent 默认没有完整父上下文;必须把任务背景、已知事实、已排除内容、关键路径、文件、日志、命令、限制和用户原话打包进去。
- 用户通常只会说功能、感觉和不满意点;brief 必须把“高级、简单、别花、像人写的、照图做、优化一下”翻译成可执行假设和验收标准。
- 描述字段决定路由质量:写清何时使用、何时不用、触发词、反例和交接边界;不要只写“擅长分析问题”。
- 权限先于能力:只读、可执行、可编辑、外部写、生产写必须分层;写操作、DB、发布、发消息、改权限、付费和删除默认需要明确授权。
- 输出必须可验收:结论要绑定 file:line、命令输出、URL 原文、测试结果、截图、API 样本或“未验证”声明。
- 停止条件必须前置:证据不足、影响面扩大、高风险副作用、连续两次无效、工具权限不足、冲突指令、超预算时停止并报告。
- 不为了并行而拆 Agent:只有独立子问题、不同专业域、搜索空间过大或需要隔离上下文时才拆;过碎分工会丢失语境。
- 不能让子 Agent 接管最终判断:子 Agent 负责收集、验证、局部实现或第二意见;主流程负责综合、取舍和最终交付。
- 每个 brief 都要写回传格式;如果不指定,结果会变成散文,难以合并、审计和复验。
口语需求归一化
- brief 中保留 user_quote,并新增 normalized_intent;normalized_intent 只写可执行解释,不写空泛审美词。
- “优化一下 / 改好看点”:归一成目标用户、当前问题、改动范围、完成状态、验收截图或命令。
- “高级 / 现代 / 别土”:归一成低噪声、强对齐、真实组件、少颜色、合适留白、明确主操作和可扫描层级。
- “简单 / 极简 / 浅色”:归一成浅色画布、白/浅灰 surface、细边框、少阴影、1 个主色、最多 3 个主导色族。
- “看图做 / 复刻”:归一成先抽 layout DSL,再实现;证据不足标 confidence,不要求子 Agent 先写长 brief。
- “生图 / 参考图”:归一成画面用途、主体、layout、色彩预算、negative prompt、比例和审图清单;默认浅色极简 UI mockup,除非用户明确要强视觉。
- 缺非关键细节时写默认假设并继续;缺平台、权限、真实品牌/人物/IP、生产写或高风险合规时才停止询问。
可复制 brief 片段:
user_quote: "{用户原话}"
normalized_intent: "{把口语需求转成具体目标、范围、视觉/质量标准}"
default_assumptions:
- UI/image style defaults to light minimal, neutral surfaces, one subdued primary color, max 3 dominant color families.
- Missing non-critical details are assumptions, not blockers.
acceptance:
- Output is directly usable or clearly marked unverified.
- Visual result avoids unnecessary color noise and follows the requested function first.
Agent brief 必填字段
- 任务标题:一句话说明本次委派要解决的问题。
- Agent 类型:Explore、Plan、general-purpose、测试、审计、研究或远程执行;说明选择理由。
- 任务类型:只读、可运行命令、可修改文件、可调用外部 API、可部署;必须和权限匹配。
- 背景:用户原始目标、业务目的、当前进度、已知事实、已排除假设、相关约束。
- 目标:本次 Agent 必须回答或完成的最小闭环,不写泛化目标。
- 非目标:明确不要改什么、不要查什么、不要扩展什么、不要凭什么下结论。
- 输入上下文:路径、函数、接口、日志、错误、截图、文档、URL、命令、环境、版本、账号地域和样本。
- 搜索范围:允许查看的目录、关键词、API、外部来源和时间范围;缺范围时按只读探索处理。
- 工具权限:允许工具、禁止工具、需确认工具、最大命令/查询/轮次数。
- 副作用边界:文件写入、DB 写入、生产 API、部署、发通知、创建 PR/issue、删除资源的确认规则。
- 执行步骤:建议路径、分叉判断、证据收集顺序、验证方式。
- 证据要求:每个结论绑定来源;本地代码用 file:line,外部资料用标题、URL、日期和适用范围。
- 停止条件:完成条件、阻塞条件、风险升级条件、预算/时间/轮次上限。
- 输出格式:结论、证据、风险、缺口、建议、下一步;必要时给表格字段。
- 验收标准:主流程如何判断这个 Agent 的结果可用、不可用或需复查。
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 7d ago First seen · 480 lines · 105 tokens per session scan A 21741994a937
agent-briefing is a skill published in the GitHub repository cass-2003/local-workflow-skill (12 stars, last pushed 2mo ago), licensed MIT. It adds 105 tokens to every session and 12,234 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
next-partial-prefetching-adoption
Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…
chronicle
Analyze Copilot session history for standup reports, usage tips, session search, and session reindexing. Use when the user asks for a standup, daily summary, usage tips, workflow recommendations, wants to search or find past sessions by keyword/file/PR, wants to reindex their session store, or asks about deleting…
babysit-pr
Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…