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/alsoleg89/doubt/agents-mdgit clone --depth 1 https://github.com/alsoleg89/doubtWhat 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 | $0.00257 | $0.00257 |
| Opus 5 | $0.00129 | $0.00129 |
| Sonnet 5 | $0.00051 | $0.00051 |
| Haiku 4.5 | $0.00026 | $0.00026 |
Grade A, and why
doubt 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 2d 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.
What it actually says
Contributor guide
Scope
Doubt is a zero-dependency Node.js CLI plus a portable Agent Skill. Keep the core local-first, provider-independent, and fail-closed on source traceability.
Commands
npm test— run the Node test suite.npm run lint— syntax-check executable source files.npm run check— run all release checks.npm run demo— render the public dogfood evidence map.node bin/doubt.js validate examples/agent-skills-vs-mcp.doubt.json— inspect the canonical example.
Design constraints
- Support Node.js 18 and later with built-in modules only.
- Never execute installed skill or map content.
- Never upload user text, source content, or telemetry. Only the explicit
doubt verifycommand may retrieve recorded source URLs; all other commands stay offline andverifymust fail closed. - Treat
skill/doubt/as the canonical install payload. - Escape all user-controlled text before embedding it in HTML.
- Add passing and adversarial fixtures for every schema invariant.
- Prefer a specific unknown over a confidence percentage.
- Never turn topical similarity into a
supportsedge.
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.
- 2d ago First seen · 29 lines · 257 tokens per session scan A 5ffa60abc17f
doubt AGENTS.md is an instructions file published in the GitHub repository alsoleg89/doubt (5 stars, last pushed 1mo ago), licensed MIT. It adds 257 tokens to every session, about $0.0013 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
council-of-high-intelligence CLAUDE.md
Instructions for 0xNyk/council-of-high-intelligence, covering council of high intelligence, architecture, conventions, agent files and skill.md.
cc-thinking-skills CLAUDE.md
Instructions for tjboudreaux/cc-thinking-skills: Follow AGENTS.md for repository workflow.
detroit-ai-player AGENTS.md
Instructions for Baba88611/detroit-ai-player, covering agents.md — 给 ai agent 的部署与运行指引, 这是什么, 前置检查, 装依赖 and 选后端（决策树，按顺序判断）.
detroit-ai-player CLAUDE.md
Instructions for Baba88611/detroit-ai-player, covering detroit: become human — ai 自主决策实验, 项目目标, 实验概述, 项目结构 and 阶段依赖与推进顺序.
life_OS AGENTS.md
Instructions for jasonhnd/life_OS, covering life os · repository guidance, repository scope, authority and layers, development principles and file changes.
foreman CLAUDE.md
Instructions for fatihguner/foreman, covering foreman — project context, architecture overview, system layers, core content layers and orchestration layers.