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 agentmods add instructions/hao-cyber/session-map/agents-mdgit clone --depth 1 https://github.com/hao-cyber/session-mapWhat 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 | $0.02889 | $0.02889 |
| Opus 5 | $0.01444 | $0.01444 |
| Sonnet 5 | $0.00578 | $0.00578 |
| Haiku 4.5 | $0.00289 | $0.00289 |
Grade A, and why
session-map 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 yesterday.
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 — 139 lines — stays where its author put it; the contents beside it link to each section on GitHub.
SessionMap 项目指令
产品事实源
- 修改产品行为、UI、信息架构、命名、持久化、运行时或分发之前，必须完整阅读
docs/product-design.md。 - 该文档是产品与设计宪章。功能规格、设计稿、框架惯例或实现便利与它冲突时， 以其中的裁决优先级为准。
- 不变的结果目标是：用户切回任何一条工作线时，能在三秒内找回此前的思考位置。
模块文档索引
- 仓库目录、workspace 与构建边界：
docs/repository-structure.md - 全局组件关系、写入协议与安全边界：
docs/architecture.md - 状态模型、轨迹、快照与树操作：
docs/modules/state-model.md - transcript 发现、增量采集与模型滚动：
docs/modules/ingestion.md - session 关联、终端状态、切换与恢复：
docs/modules/session-navigation.md - 主题 → session → 脉络的 Web 读取层：
docs/modules/web-ui.md - 本地 Web 服务、浏览器入口与 launchd：
docs/modules/local-runtime.md - 安装式 Web App、macOS 极薄壳与桌面召回：
docs/modules/desktop-host.md - 旧状态迁移、standalone CLI 与发布：
docs/modules/migration-release.md
先按改动范围读取对应模块文档；跨模块改动同时读取所有受影响文档。模块文档解释 “代码现在如何守住契约”，不得覆盖或重新解释产品宪章。
问题求解与文档演进
- 遇到卡点、反复失败或问题边界不清时，先整理已知事实、约束、假设和未知量，再把 具体故障抽象成若干可检索的问题。确认它是否已有成熟的问题名称、理论模型、标准 解法或业界实践；优先查阅一手资料和成熟实现，并明确外部解法与本项目约束的映射， 不得未经验证直接照搬。
- 修改模块前必须阅读对应模块文档；改动使现有职责、契约、数据流、边界或验证方式 发生变化时，功能与相关文档必须在同一改动中更新。文档描述当前真实设计，不把 changelog 混入正文；实现细节不得反向篡改产品宪章。
- 文档随模块增长而变得过长、混合多个可独立理解的子主题，或单文件已难以可靠导航时， 按 hub-and-spoke 方式拆分：hub 保留职责、核心不变量、全局关系和 spoke 索引，spoke 承载具体协议、流程或子模块细节。Hub 与 spoke 必须双向链接；新增一级模块边界时同步 更新本文件的模块文档索引，避免产生不可发现的孤立文档。
架构取向与变更门禁
-
追求简洁、显式、可推导的模块化单体：职责高内聚，依赖沿公开契约单向流动；同一 事实只有一个来源，同一类副作用只有一个所有者。能减少概念、状态、入口和机制时， 不增加抽象。
-
新能力必须沿既有数据流和模块边界接入。禁止“飞线”：跨层调用内部实现、绕过统一 写入、校验、鉴权或生命周期，复制持久状态，引入旁路入口，或用特例把本应独立的 模块绑在一起。
-
若合理需求反复需要越界，先修正模块职责或公开契约，并同步架构文档；不得用无归属 的 helper、共享可变状态或重复编排掩盖边界错误。抽象只用于消除已经存在的重复或 隔离已确认的变化，不为假设中的未来扩展预建层级。
-
改动跨越模块边界，或改变职责、数据所有权、写入路径、信任边界、生命周期、失败恢复 与分发方式时，编码前先完成轻量架构审查。
-
审查至少明确：受影响契约、唯一事实源与写入者、数据流和依赖方向、失败与重启行为、 安全边界、迁移与回滚方式，以及能够证明这些约束的验证证据。
-
重大架构决策必须记录背景、约束、备选方案、取舍和后果；后续推翻时标记替代关系， 不得静默改写历史决定。
-
删除旧实现、旧状态或兼容桥之前，必须证明新路径完成真实迁移、重启恢复、回滚和安装 冒烟；“新实现能运行”不等于“旧路径可以清理”。
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.
- yesterday First seen · 139 lines · 2,889 tokens per session scan A 1b085207062e
session-map AGENTS.md is an instructions file published in the GitHub repository hao-cyber/session-map (2 stars, last pushed 18d ago), licensed MIT. It adds 2,889 tokens to every session, about $0.0144 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 2026-08-31.
Other instructions, from other repositories
codex-provider-sync AGENTS.md
Instructions for Dailin521/codex-provider-sync, covering ai / agent operator guide, vnext architecture baseline, goal, choose the interface and safe operating flow.
munkel CLAUDE.md
Instructions for limehq/munkel, covering claude.md, monorepo layout, commands, macos development specifics and deploy.
codex-studio AGENTS.md
Instructions for gvastethecreator/codex-studio, covering codex studio agent guide, current direction, setup / first run, required context pass and commands.
agentpulse AGENTS.md
Instructions for jstuart0/agentpulse, covering agents.md - agentpulse, tech stack, commands, project structure and architecture.
tokenjuice AGENTS.md
Instructions for kendrick-na/tokenjuice, covering agents.md — installing tokenjuice, step 0 — is it already installed?, step 1 — detect the os, macos and windows.
codex-token-meter AGENTS.md
Instructions for prefect12/codex-token-meter, covering ai token meter development guide, project shape, build and verification, git workflow and ai token meter test loop.