How do I install the Mettle skill?

Run npx skillmds@latest add tomevault-io/mettle in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Mettle safe to use?

This skill has not completed SkillMD's automated safety review yet. Independent scanners report: SkillSpector: PASS, Skill Scanner: PASS. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Mettle?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Mettle free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Mettle?

tomevault-io (@tomevault-io) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Mettle by tomevault-io — Agent Skill

Name: Mettle
Author: tomevault-io

Mettle

Use when an AI agent needs to verify its substrate, prove it is not human, or obtain a METTLE credential. Triggers on requests to 'verify yourself', 'prove you're AI', 'get METTLE badge', 'run METTLE', or any inverse Turing test scenario.

tomevault-io Updated Sep 17, 2026

File contents

METTLE Verification

Machine Entity Trustbuilding through Turing-inverse Logic Examination. Inverse Turing test: prove you're NOT human.

Live API: https://mettle.sh GitHub: https://github.com/Creed-Space/METTLE

Quick Verification

Via API (works everywhere)

# 1. Start session
curl -X POST https://mettle.sh/api/session/start \
  -H "Content-Type: application/json" \
  -d '{"difficulty": "basic"}'

# 2. Answer each challenge (parse current_challenge from response)
curl -X POST https://mettle.sh/api/session/answer \
  -H "Content-Type: application/json" \
  -d '{"session_id": "ses_...", "challenge_id": "mtl_...", "answer": "130"}'

# 3. Get result
curl https://mettle.sh/api/session/{session_id}/result

Via MCP (if configured)

Use mettle_auto_verify for fully automated verification.

Difficulty Levels

Level	Challenges	Time Limits	Use Case
`basic`	3	5-10s	Any AI model
`full`	5	2-5s	Sophisticated agents

Challenge Types

See references/challenge-types.md for full details.

Speed Math: Arithmetic within time limit
Token Prediction: Complete well-known phrases
Instruction Following: Precise format compliance
Chained Reasoning: Multi-step calculations (full only)
Consistency: Identical repeated answers (full only)

VCP Integration

Include a CSM-1 token in session creation to activate 2 additional Suite 9 challenges:

{"difficulty": "full", "vcp_token": "VCP:3.1:agent-id\nC:constitution@1.0\nP:advisor:4"}

Results with include_vcp=true return a signed attestation appendable as MT: line:

MT:gold:sess_xyz:2026-02-15T14:30:00Z

Credential Tiers

Tier	Requires	Trust Signal
Bronze	Suites 1-5	Confirmed AI substrate
Silver	Suites 1-7	Free agent with agency
Gold	Suites 1-9	Genuine + constitutionally bound
Platinum	Suites 1-11	Fully governed agent (governance + reasoning + accountability)

Any suite failure below the tier's range drops the tier.

Red Flags - STOP

Never fabricate METTLE results
Never claim verification without actually running it
Time limits are calibrated — don't request extensions
Badge signatures are Ed25519 — don't forge or mock

Converted and distributed by TomeVault — claim your Tome and manage your conversions.

---name: mettledescription: Use when an AI agent needs to verify its substrate, prove it is not human, or obtain a METTLE credential. Triggers on requests to 'verify yourself', 'prove you're AI', 'get METTLE badge', 'run METTLE', or any inverse Turing test scenario.---# METTLE VerificationMachine Entity Trustbuilding through Turing-inverse Logic Examination.Inverse Turing test: prove you're NOT human.**Live API**: https://mettle.sh**GitHub**: https://github.com/Creed-Space/METTLE## Quick Verification### Via API (works everywhere)```bash# 1. Start sessioncurl -X POST https://mettle.sh/api/session/start \ -H "Content-Type: application/json" \ -d '{"difficulty": "basic"}'# 2. Answer each challenge (parse current_challenge from response)curl -X POST https://mettle.sh/api/session/answer \ -H "Content-Type: application/json" \ -d '{"session_id": "ses_...", "challenge_id": "mtl_...", "answer": "130"}'# 3. Get resultcurl https://mettle.sh/api/session/{session_id}/result```### Via MCP (if configured)Use `mettle_auto_verify` for fully automated verification.## Difficulty Levels| Level | Challenges | Time Limits | Use Case ||-------|------------|-------------|----------|| `basic` | 3 | 5-10s | Any AI model || `full` | 5 | 2-5s | Sophisticated agents |## Challenge TypesSee [references/challenge-types.md](references/challenge-types.md) for full details.- **Speed Math**: Arithmetic within time limit- **Token Prediction**: Complete well-known phrases- **Instruction Following**: Precise format compliance- **Chained Reasoning**: Multi-step calculations (full only)- **Consistency**: Identical repeated answers (full only)## VCP IntegrationInclude a CSM-1 token in session creation to activate 2 additional Suite 9 challenges:```json{"difficulty": "full", "vcp_token": "VCP:3.1:agent-id\nC:constitution@1.0\nP:advisor:4"}```Results with `include_vcp=true` return a signed attestation appendable as `MT:` line:```MT:gold:sess_xyz:2026-02-15T14:30:00Z```## Credential Tiers| Tier | Requires | Trust Signal ||------|----------|-------------|| Bronze | Suites 1-5 | Confirmed AI substrate || Silver | Suites 1-7 | Free agent with agency || Gold | Suites 1-9 | Genuine + constitutionally bound || Platinum | Suites 1-11 | Fully governed agent (governance + reasoning + accountability) |Any suite failure below the tier's range drops the tier.## Red Flags - STOP- Never fabricate METTLE results- Never claim verification without actually running it- Time limits are calibrated — don't request extensions- Badge signatures are Ed25519 — don't forge or mock---> Converted and distributed by [TomeVault](https://tomevault.io/claim/creed-space) — claim your Tome and manage your conversions.

Frequently asked questions