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 companion-inc/introspect --skill agent-md-creatorgit clone --depth 1 https://github.com/companion-inc/introspectWrote 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/companion-inc/introspect/agent-md-creator)<a href="https://agentmods.dev/skills/companion-inc/introspect/agent-md-creator"><img src="https://agentmods.dev/badge/skills/companion-inc/introspect/agent-md-creator/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/skills/companion-inc/introspect/agent-md-creator"><img src="https://agentmods.dev/badge/skills/companion-inc/introspect/agent-md-creator.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 | $0.00060 | $0.02064 |
| Opus 5 | $0.00030 | $0.01032 |
| Sonnet 5 | $0.00012 | $0.00413 |
| Haiku 4.5 | $0.00006 | $0.00206 |
Grade A, and why
agent-md-creator 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 10d 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 — 77 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent MD creator
The one truth everything follows from
A prompt is soft conditioning on a probability distribution, not control. An LLM is a next-token predictor trained by imitation (so it defaults to the modal / consensus continuation), then RLHF'd to maximize a proxy for human approval (so it agrees, looks-done, and capitulates under pushback), running stateless with no reliable metacognitive monitor (so it cannot feel its own memory or limits). Every common agent failure is one of these three facts seen from a different angle. Therefore a prompt can raise the probability of a behavior but never guarantee it, and adherence decays as the file grows. Design around this; don't fight it.
Layer map — put each rule where it can actually work
- Prompt (advisory): judgment, approach, taste — "what good looks like." Followed most of the time, not all.
- Hooks / harness (deterministic): non-negotiables that must happen every time. The only layer outside the sampled token stream, so the only one that enforces.
- Skills (on-demand): workflows relevant only sometimes — keep them out of the always-loaded prompt.
- Training: moves the floor. Not yours to change; the frontier failures (first-principles reasoning, self-monitoring, taste) live here — no prompt line fixes them.
If a rule must hold 100% of the time, it is a hook, not a prompt line.
Scope map — global, project, override, local
- Codex global:
~/.codex/AGENTS.md, or~/.codex/AGENTS.override.mdwhen the entire global file must be replaced. Use only for cross-project invariants. - Codex project:
AGENTS.mdin the repo root or nested directory. Codex concatenates global, repo, and nested files; later/closer files win on conflicts. - Codex override:
AGENTS.override.mdreplaces the regularAGENTS.mdfor that directory level. Use it only for a subtree that must not inherit that level's normal guidance. - Claude project: Claude reads
CLAUDE.md, notAGENTS.md; createCLAUDE.mdwith@AGENTS.mdwhen shared repo guidance should load in Claude, then append Claude-only additions below it. - Claude local:
CLAUDE.local.mdis private project-specific guidance and should be gitignored. - Project skills: Codex repo skills live under
.agents/skills/<skill>/SKILL.md; Claude project skills live under.claude/skills/<skill>/SKILL.md. Use them for codebase-specific workflows that should not follow the user everywhere.
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.
- 10d ago First seen · 77 lines · 60 tokens per session scan A 952c5aae421e
agent-md-creator is a skill published in the GitHub repository companion-inc/introspect (10 stars, last pushed 21d ago), licensed MIT. It adds 60 tokens to every session and 2,064 once invoked, about $0.0003 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 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…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…
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…