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/mrtoyy/convosketchpad/agents-mdgit clone --depth 1 https://github.com/MrToyy/convosketchpadWrote 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/mrtoyy/convosketchpad/agents-md)<a href="https://agentmods.dev/instructions/mrtoyy/convosketchpad/agents-md"><img src="https://agentmods.dev/badge/instructions/mrtoyy/convosketchpad/agents-md.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 | $0.02073 | $0.02073 |
| Opus 5 | $0.01037 | $0.01037 |
| Sonnet 5 | $0.00415 | $0.00415 |
| Haiku 4.5 | $0.00207 | $0.00207 |
Grade A, and why
convosketchpad 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 4d 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 — 89 lines — stays where its author put it; the contents beside it link to each section on GitHub.
仓库 Agent 指南
本仓库是独立维护的 ConvoSketchpad，包含 Canvas、Agent 运行端集成与受管用户认证。开始工作前必须先阅读本文，再按任务索引读取对应文档；不要依靠全仓搜索重新推断已有设计。
如果以后在子目录增加更具体的 AGENTS.md，以距离目标文件最近的规则为准。用户在当前任务中的明确要求优先于仓库约定。
产品与职责边界
- ConvoSketchpad 不是独立的 Agent 运行环境，必须连接到至少一个可用的 Agent 运行端（Agent Runtime）；当前 Release 支持 OpenClaw Gateway 与本地 Codex App Server，统一边界见
docs/ARCHITECTURE.md和docs/agent-runtimes/ADAPTER-DEVELOPMENT.md。 - 所选 Agent 运行端负责 Agent、工具、执行、Conversation、事件和原始运行记录；Codex 通过受监督的本地 App Server 直接接入，不经过 OpenClaw。
- ConvoSketchpad 负责 Canvas 拓扑与布局、Branch/Interaction 关系、发送协调、恢复元数据、受管用户隔离，以及附件和 Artifact 的持久化副本。
- 当前 OpenClaw 能力优先使用原生 Gateway 或 CLI 接口，不得新增对 OpenClaw 本地配置文件或 Agent 工作区的直接读写依赖；通用 Canvas 代码不得直接依赖 Runtime 协议方法名。
- 直接 Codex 集成只能通过受监督的 App Server 协议和明确的统一 Agent Runtime Port；不得检查 Codex/Claude 本地凭据文件，也不得调用它们的 CLI 获取 Provider、用量或配额数据。
任务索引
| 任务 | 首先阅读 | 说明 |
|---|---|---|
| 不确定从哪里开始、查看完整文档 | docs/README.md |
全部产品、运维和维护者文档索引 |
| 产品目标、体验或职责边界 | docs/PRODUCT.md |
产品原则和 OpenClaw/ConvoSketchpad 分工 |
| 架构、Session、数据流或持久化 | docs/ARCHITECTURE.md |
运行时模型、发送、恢复和文件流 |
| Agent Runtime 抽象、OpenClaw/Codex 边界 | docs/ARCHITECTURE.md、docs/agent-runtimes/ADAPTER-DEVELOPMENT.md |
统一 Port、事件/审批、恢复、Artifact 与扩展约束 |
| 新增或修改 Agent Runtime Adapter | docs/agent-runtimes/ADAPTER-DEVELOPMENT.md |
目录、Port、聚合、审批、迁移、安全与测试规范 |
| Canvas 页面、节点、布局、Branch、Interaction、附件或 Artifact | docs/canvas/CANVAS-CODE-MAP.md |
Canvas 不变量、前后端入口与测试 |
| 资源引用、分支合并或跨 Agent 交接 | docs/ARCHITECTURE.md、docs/canvas/BRANCH-MERGING.md、docs/canvas/CANVAS-CODE-MAP.md |
现行数据语义、分支合并设计、验收条件和 Canvas 约束 |
| 登录、用户 Token、隔离、限流、Cookie 或 SSE 鉴权 | docs/canvas/AUTH-CODE-MAP.md |
认证链路、CLI、SQLite 与会话撤销 |
| HTTP 路由或接口契约 | docs/API.md |
Canvas、认证、文件和遥测 API |
| 本地安装、开发环境或贡献流程 | docs/INSTALL.md、CONTRIBUTING.md |
安装向导、统一开发命令和实现要求 |
| 环境变量或 Gateway 配置 | docs/CONFIGURATION.md |
配置来源、默认值和原生 OpenClaw 设置 |
| 部署、网络暴露或远程访问 | docs/DEPLOYMENT.md、docs/SECURITY.md |
部署拓扑、安全基线和信任模型 |
| 故障诊断 | docs/TROUBLESHOOTING.md |
连接、Agent、Session、文件和认证问题 |
| 更新、Release 或版本记录 | docs/UPDATING.md、CHANGELOG.md |
更新器、发布规则和变更记录 |
| 分支、Remote、提交或推送 | docs/canvas/GIT-WORKFLOW.md |
main 分支职责和 Git 安全步骤 |
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.
- 4d ago First seen · 89 lines · 2,073 tokens per session scan A 255acbf2e3d4
convosketchpad AGENTS.md is an instructions file published in the GitHub repository MrToyy/convosketchpad (1 stars, last pushed 22d ago), licensed MIT. It adds 2,073 tokens to every session, about $0.0104 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
SmolVM AGENTS.md
Instructions for CelestoAI/SmolVM, covering smolvm context, 🚀 project overview, 🧪 development, key commands and release checklist.
SmolVM CLAUDE.md
Instructions for CelestoAI/SmolVM, a project described as: Open-source AI sandbox infrastructure with unified API for VMMs -- Firecracker, QEMU and libkrun.
Sponsio CLAUDE.md
Claude Code instructions for SponsioLabs/Sponsio, covering agent guide for sponsio, what sponsio is, positioning, what to read first and repository map.
librefang AGENTS.md
AGENTS.md instructions for librefang/librefang, covering stack, layout, build, architecture and kernelhandle trait.
gini-agent AGENTS.md
AGENTS.md instructions for Open-Curiosity/gini-agent, covering gini agent instructions, shape, adrs, boundaries and branches.
autonomous-os AGENTS.md
AGENTS.md instructions for autonomous-ai/autonomous-os, covering agents.md, multi-ide rules, working style, parallel work / subagents and device access rules.