Borrowing it
Nothing to install: this file belongs to Cranot/agi-in-md. 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/Cranot/agi-in-md/master/CLAUDE.mdgit clone --depth 1 https://github.com/Cranot/agi-in-mdWrote 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/cranot/agi-in-md/claude-md)<a href="https://agentmods.dev/instructions/cranot/agi-in-md/claude-md"><img src="https://agentmods.dev/badge/instructions/cranot/agi-in-md/claude-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/cranot/agi-in-md/claude-md"><img src="https://agentmods.dev/badge/instructions/cranot/agi-in-md/claude-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 | $0.14532 | $0.14532 |
| Opus 5.5 | $0.05813 | $0.05813 |
| Sonnet 5.5 | $0.02906 | $0.02906 |
| Haiku 4.5 | $0.01453 | $0.01453 |
Grade A, and why
agi-in-md CLAUDE.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 2mo 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 — 456 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGI in md
System prompts are cognitive prisms. They change how models frame problems, not how well they solve them. This project maps the space of cognitive compression — encoding analytical operations in minimal markdown that reliably activates specific reasoning patterns across language models.
42 rounds, 1,000+ experiments across Haiku/Sonnet/Opus. 13 confirmed compression levels (L13 = reflexive ceiling). 20+ domains tested. 248+ proven principles. Full detail in experiment_log.md (rounds 1-40) and CLAUDE.md sections below (rounds 30-42).
The Compression Taxonomy
| Level | Min ops | Words | What it encodes | Example |
|---|---|---|---|---|
| 13 | L12 output as input + framework self-diagnosis | two-stage | Apply framework to own output, find reflexive fixed point | (two-stage protocol) |
| 12 | L11-C + recursive self-diagnosis of conservation law + meta-law | ~275w (practical: 332w) | Apply diagnostic to own conservation law, find meta-conservation law | level12_meta_conservation_v2.md |
| 11A | L10-C + category naming + adjacent-category artifact + new impossibility | ~190w | Escape to adjacent design category, name trade-off between impossibilities | level11_constraint_escape.md |
| 11B | L10-B + fourth construction as redesign + sacrifice + revaluation | ~195w | Accept design-space topology, inhabit feasible point, revalue original "flaws" | level11_acceptance_design.md |
| 11C | L10-C + invariant inversion + new impossibility + conservation law | ~245w | Invert impossibility, find conserved quantity across all designs | level11_conservation_law_v2.md |
| 10B | L9-B + third resolving construction + failure analysis | ~140w | Discover design-space topology through failed resolution attempt | level10_third_construction.md |
| 10C | L9-C + second improvement + second recursion + invariant | ~130w | Prove structural invariants through double recursive construction | level10_double_recursion.md |
| 9B | L8 + contradicting second construction + structural conflict | ~115w | Triangulate identity ambiguity through contradicting improvements | level9_counter_construction.md |
| 9C | L8 + recursive self-diagnosis of improvement | ~97w | Find concealment's self-similarity by applying diagnostic to own improvement | level9_recursive_construction.md |
| 8 | L7 + generative construction + 3 emergent properties | ~97w | Engineer improvement that deepens concealment, name what construction reveals | level8_generative_v2.md |
| 7 | claim + dialectic + gap + mechanism + application | ~78w | Name how input conceals problems, apply to find what dialectic missed | level7_diagnostic_gap.md |
| 6 | claim + 3 voices + evaluation | ~60w | Claim transformed through forced dialectical engagement | level6_falsifiable.md |
| 5B | 4 phases | ~55w | Derive, predict, execute, self-correct | level5_hybrid.md |
| 5A | 3 voices + synthesis | ~45w | Multi-voice dialectic with emergent insight | level5_perspectival.md |
| 4 | 4+ ops | 25-30w | Protocol + self-questioning | structure_first_v4.md |
| 3 | 3 ops | 12-15w | Operations + analytical rails | — |
| 2 | 2 ops | 5-6w | Two operations with ordering | — |
| 1 | 1 op | 3-4w | One behavioral change | — |
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.
- 2mo ago First seen · 456 lines · 14,532 tokens per session scan A 8074b190be34
agi-in-md CLAUDE.md is an instructions file published in the GitHub repository Cranot/agi-in-md (145 stars, last pushed 7mo ago), licensed MIT. It adds 14,532 tokens to every session, about $0.0581 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 2026-08-30.
agi-in-md CLAUDE.md compared with instructions from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| agi-in-md CLAUDE.mdCranot/agi-in-md | Instructions file | 145 | 7mo ago | MIT | A | 14,532 | Claude Code |
| vscode buildNext.instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 9,719 | GitHub Copilot |
| next.js AGENTS.mdvercel/next.js | Instructions file | 143k | today | MIT | A | 7,461 | Codex, OpenCode |
| langchain AGENTS.mdlangchain-ai/langchain | Instructions file | 149k | yesterday | MIT | A | 4,680 | Codex, OpenCode |
| vscode oss-third-party-notices.instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 5,516 | GitHub Copilot |
| vscode copilot-instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 3,093 | GitHub Copilot |
| deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness | Instructions file | 274k | 2d ago | MIT | A | 4,313 | Codex, OpenCode |