Borrowing it
Nothing to install: this file belongs to Stahl-G/briefloop. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/Stahl-G/briefloop/main/AGENTS.mdgit clone --depth 1 https://github.com/Stahl-G/briefloopWrote 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/instructions/stahl-g/briefloop/agents-md)<a href="https://agentmods.dev/instructions/stahl-g/briefloop/agents-md"><img src="https://agentmods.dev/badge/instructions/stahl-g/briefloop/agents-md/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/instructions/stahl-g/briefloop/agents-md"><img src="https://agentmods.dev/badge/instructions/stahl-g/briefloop/agents-md.svg" alt="Reviewed on agentmods" width="80" 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 | $0.01957 | $0.01957 |
| Opus 5.5 | $0.00783 | $0.00783 |
| Sonnet 5.5 | $0.00391 | $0.00391 |
| Haiku 4.5 | $0.00196 | $0.00196 |
Grade A, and why
briefloop AGENTS.md 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 26d 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 — 55 lines — stays where its author put it; the contents beside it link to each section on GitHub.
BriefLoop 开发约定
这是本地研究与报告工具,产出企业内部可读、可核查的报告。界面上的主 Agent 叫 BriefLoop:负责对话、研究规划、写稿与修订,模型用主链配置;Evaluator、Wiki Maintainer、Skill Proposer 三者独立运行,可分别指定模型。目标:重要结论可追溯,草稿持续可编辑,正式交付与独立审阅和证据版本绑定,通过既有 WikiSkill 学习已核实的修订。
- 若存在 private_planning/PLAN.md,以其中的最新用户决定为准;普通 clone 以本文件和 README 为准。旧项目的阶段、审批和测试约定不自动继承。
- 实现真实可操作的网页和后台 agent 执行,不把手动 JSON 或 mock 当成完成。
- 前端新增功能直接写进
frontend/下的新模块(工厂函数注入依赖,参考 schedules.js、app-updates.js 的写法);app.js只保留全局状态、启动与组装,不再整块增长。 - 保留精简 SQLite ControlStore、来源、稿件版本、评分绑定和恢复;不复制旧 mandatory Auditor/Editor/Claim Ledger 流程。
- BriefLoop(主 Agent,代码中也叫 Orchestrator)和实际 agent 负责规划、研究、写作、评分;Python 负责工具、进程、存储、确定计算及既定选择规则。
- 用户原稿/修订可见;低分不隐藏稿件,普通工作稿始终可下载。正式交付不能豁免确定重大错误、核心证据缺口或未决重要冲突。Wiki 是经验,不是当期事实来源。
- Reviewer 独立只读核对实际产物、原件与本任务历史,不自行重跑分析、补搜或改稿;权限由选定宿主实际落实,缺能力不得用提示词冒充隔离。
- 复用现有 jobs/events/Store 和一次自动修订;来源正常更新与原稿错误分开。审计包由用户单独选择导出,不含凭据和隐藏推理,含各材料的定位与校验结果。
- 默认 Scout/Analyst 可演化,但角色/技能/评价接口可扩展。K 在报告反馈页调整,默认 1。
- 只保留必要测试,不追求覆盖率、千级旧回归或状态组合矩阵;细则见下方“测试”。
- 哈希只用在内容会在别处变化、且变化要被发现的地方:来源与稿件版本绑定、下载与发行工件、导出输入、交给 agent 可写目录的文件、审计包和缓存键。不给只由 BriefLoop 写入 SQLite 的记录再存一份自身哈希并在读取时核对;那只能防住不会发生的事,也拦不住真正能改库的人。
- 不覆盖旧仓库和用户改动;外部代码保留许可证与来源说明。
- 凭据/运行数据/详细计划不进 Git。公共安装不能依赖个人绝对源码路径。
- 普通本地实现及必要修复已获授权,不逐步请求批准;外发、发布和新付费服务另计。
代码与文案
- 同一限制只在最相关的位置写一次。README、文档和界面只写用户必须知道的限制,不逐段追加“不代表/不等于/不证明”。
- 提示词写角色目的、判断标准和交付物。工具参数、字段格式和取值范围写在工具定义或 schema 里,由程序校验,不在提示词里重复;能用肯定句说清的不写成禁止句。
- 兼容只针对已发布版本写入过的数据(以发布标签为准)。开发中、未发布的格式直接改,不留 legacy 分支、版本常量或对应测试。
- 代码单行不超过约 160 字符(提示词和界面文案的正文段落除外),不用分号把一长串语句塞进一行;SQL 分行写。新增函数超过约 150 行时拆开。
frontend/app.js不再加功能,见上文。前端测试取函数用tests/source_section.mjs的section/statement,不要假设一个函数只占一行。
测试
测试数量不是交付证据。2026-10 已删除约 1,500 个只复述实现的测试,不要再写回来。
- 写测试前先问:实现改错了,这个测试会失败吗? 如果只在改措辞、改常量、改写法时才失败,就不写。
- 不写:
- 断言提示词、方法卡、文档、界面文案里含某句话;
- 把代码里的常量、映射、默认值原样抄进测试再比一遍;
- 读
app.js、index.html、CSS 或 Python 源码做字符串或正则匹配; - 给
experiments/、native-engine/experiment_*和一次性脚本写测试; - 同一保证换几种并发、旧格式或取消时机各写一个;只留最能复现问题的那一个。
- 只为这些行为写测试:保存与版本绑定、失败恢复、选择与回退、权限与沙箱边界、凭据脱敏、引用与来源完整性、确定计算的边界情况、导出往返、Windows 等平台路径。前端测试要真正执行模块函数,不读源码文本。
- 改提示词、方法卡、文档或界面文案不需要新测试;用一次实际运行或生成结果检查。
- 新增测试前先找现有测试,能扩一个断言就不另起一个。一个 PR 新增测试通常不超过 2 个;超过时在 PR 说明里写清每个测试防的是什么错误。
- 修 review 意见时,只有问题能用真实输入复现,才补回归测试;“防止以后有人改回去”不是理由。
- 跑相关测试文件即可,不必每次跑全量;最终做一次小型真实闭环。
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.
- 26d ago Changed · +25 lines · +752 tokens per session 2f5cb25c3fe6
- 1mo ago Changed · +2 lines · +93 tokens per session ade712e94cc9
- 1mo ago Changed · +1 lines · +58 tokens per session 3cfb7ec46226
- 1mo ago Changed · +10 lines · +461 tokens per session 31293dcbe2fe
- 1mo ago Changed · +2 lines · +216 tokens per session 73221317945a
- 1mo ago Changed · -206 lines · -2,128 tokens per session bfc5c772134d
- 2mo ago Changed · +9 tokens per session fc581d7c4260
- 2mo ago First seen · 221 lines · 2,496 tokens per session scan A 1d9ce50cd17c
briefloop AGENTS.md is an instructions file published in the GitHub repository Stahl-G/briefloop (9 stars, last pushed 21d ago), licensed MIT. It adds 1,957 tokens to every session, about $0.0078 per session on Opus 5.5. 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 2026-08-31.
briefloop AGENTS.md compared with instructions from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| briefloop AGENTS.mdStahl-G/briefloop | Instructions file | 9 | 21d ago | MIT | A | 1,957 | Codex, OpenCode |
| vscode buildNext.instructions.mdmicrosoft/vscode | Instructions file | 195k | today | MIT | A | 9,719 | GitHub Copilot |
| next.js AGENTS.mdvercel/next.js | Instructions file | 143k | today | MIT | A | 7,461 | Codex, OpenCode |
| vscode oss-third-party-notices.instructions.mdmicrosoft/vscode | Instructions file | 195k | today | MIT | A | 5,516 | GitHub Copilot |
| langchain AGENTS.mdlangchain-ai/langchain | Instructions file | 149k | today | MIT | A | 4,680 | Codex, OpenCode |
| vscode copilot-instructions.mdmicrosoft/vscode | Instructions file | 195k | today | MIT | A | 3,093 | GitHub Copilot |
| deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness | Instructions file | 275k | 3d ago | MIT | A | 4,313 | Codex, OpenCode |