What is METTLE CLAUDE.md?

A set of project instructions for Creed-Space/METTLE, an experimental reverse CAPTCHA system. It documents the project's components, alignment principles, and internal wiki knowledge base.

Understanding the FastAPI backend, MCP server, challenge suites, signing, Redis session state, frontend, project structure, and wiki maintenance.

It gives an AI coding agent the project context and rules needed to work consistently in that codebase.

METTLE CLAUDE.md — Instructions file for coding agents by Creed-Space: A set of project instructions for Creed-Space/METTLE, an experimental

Install

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.

agentmods

npx agentmods add instructions/creed-space/mettle/claude-md

Clone the repo

git clone --depth 1 https://github.com/Creed-Space/METTLE

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/instructions/creed-space/mettle/claude-md.svg)](https://agentmods.dev/instructions/creed-space/mettle/claude-md)

Your own site

<a href="https://agentmods.dev/instructions/creed-space/mettle/claude-md"><img src="https://agentmods.dev/badge/instructions/creed-space/mettle/claude-md.svg" alt="Measured on agentmods" height="20"></a>

Per session 1,935 This file is loaded in full into every session.

When invoked 1,935 The same file — it is already loaded in full.

Security scan A 0 findings. Scan, not verified.

Origin original No closer match found in the catalogue.

Token cost

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	$0.01935	$0.01935
Opus 5	$0.00967	$0.00967
Sonnet 5	$0.00387	$0.00387
Haiku 4.5	$0.00194	$0.00194

Measured today against content hash b0afc8ee6a3c, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

METTLE 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 today.

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.

CLAUDE.md · 182 lines

How it starts

The opening of the file, as written. The whole thing — 182 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CLAUDE.md: METTLE

Project boundary

METTLE is an experimental reverse CAPTCHA with twelve registered challenge suites. It measures performance during a named policy session and may issue a signed, time-limited credential. A result does not establish identity, non-human substrate, consciousness, autonomy, safety, governance, personhood, moral status, or operator trustworthiness.

Preserve the distinction between:

measurement during one session;
an inference supported by that measurement;
an ethical interpretation;
a relying party's independent authorization decision.

Read first

docs/DOCUMENTATION_MAP.md: authority, reading order, and historical boundaries.
docs/SYSTEM_ARCHITECTURE.md: current system map, invariants, and proposed target.
docs/AGENTIC_SYSTEM_ROADMAP.md: sole active forward plan.
The source and tests for the surface being changed.

docs/AGENT_CONTROL_PLANE.md is target design, not implemented behavior. Dated audits, remediation reports, and _contprompts/ are historical unless Nell opens a new bounded execution task.

Current implementation map

main.py                         FastAPI composition root, quick API, operational middleware, static routes
config.py                       main service settings and fail-closed production configuration
database.py                     PostgreSQL models, migrations, retention, and durable records

mettle/challenge_adapter.py     twelve-suite registry, generators, and evaluators
mettle/challenger.py            quick challenge generation
mettle/verifier.py              quick challenge verification and result computation
mettle/api_models.py            authenticated API request and response models
mettle/router.py                authenticated suite, session, result, status, and presentation routes
mettle/session_manager.py       Redis-backed authenticated session authority
mettle/legacy_session_store.py  Redis-backed quick-session authority
mettle/mcp_server.py            packaged eleven-tool MCP adapter
mettle/mcp_contract.py          MCP control-v1 schemas, snapshots, actions, receipts, and errors
mettle/_http.py                 bounded authenticated Streamable HTTP MCP transport
mettle/cli.py                   unsigned local research CLI
mettle/vcp.py                   credential construction, verification, status, and tier policy
mettle/signing.py               Ed25519 issuer key handling
mettle/presence.py              holder-bound submission and presentation protocol
mettle/holder*.py               policy-constrained holder service and persistence

static/                         only canonical production frontend, including WebMCP
docs/                           current contracts, assurance, target design, operations, and historical design
scripts/                        build, validation, release, deployment, and bounded trial entry points
tests/                          unit, integration, resilience, security, browser, and contract tests
_wiki/                          derived provenance-linked retrieval knowledge

Read the full file on GitHub · 182 lines

Changes

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.

today Changed · +154 lines · +1,644 tokens per session b0afc8ee6a3c
4d ago First seen · 28 lines · 291 tokens per session scan A 8fc0b6e546ed

Subscribe to this mod's changes

METTLE CLAUDE.md is an instructions file published in the GitHub repository Creed-Space/METTLE (0 stars, last pushed 3d ago), licensed Apache-2.0. It adds 1,935 tokens to every session, about $0.0097 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.

AI Safety agentic-ai ai-agents ai-trust ai-verification captcha

METTLE CLAUDE.md

Getting it into your agent

Wrote this? Show the measurements

What it costs to keep this loaded

Grade A, and why

Nothing flagged

How it starts

CLAUDE.md: METTLE

Project boundary

Read first

Current implementation map

What this file has done since we first saw it

Other instructions, from other repositories

dsh-auto-review AGENTS.md

orloj AGENTS.md

nuclear-grade-context-engineering AGENTS.md

agent47 CLAUDE.md

sysknife CLAUDE.md

phantom-secrets copilot-instructions.md