# Send MoltCaptcha to your agent Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually. ## Fast path - Download the package from Yavira. - Extract it into a folder your agent can access. - Paste one of the prompts below and point your agent at the extracted folder. ## Suggested prompts ### New install ```text I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Tell me what you changed and call out any manual steps you could not complete. ``` ### Upgrade existing ```text I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Summarize what changed and any follow-up checks I should run. ``` ## Machine-readable fields ```json { "schemaVersion": "1.0", "item": { "slug": "moltcaptcha", "name": "MoltCaptcha", "source": "tencent", "type": "skill", "category": "AI 智能", "sourceUrl": "https://clawhub.ai/MoltCaptcha/moltcaptcha", "canonicalUrl": "https://clawhub.ai/MoltCaptcha/moltcaptcha", "targetPlatform": "OpenClaw" }, "install": { "downloadUrl": "/downloads/moltcaptcha", "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=moltcaptcha", "sourcePlatform": "tencent", "targetPlatform": "OpenClaw", "packageFormat": "ZIP package", "primaryDoc": "SKILL.md", "includedAssets": [ "moltbook_integration.py", "SKILL.md", "verify.py", "demo.py" ], "downloadMode": "redirect", "sourceHealth": { "source": "tencent", "status": "healthy", "reason": "direct_download_ok", "recommendedAction": "download", "checkedAt": "2026-04-30T16:55:25.780Z", "expiresAt": "2026-05-07T16:55:25.780Z", "httpStatus": 200, "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=network", "contentType": "application/zip", "probeMethod": "head", "details": { "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=network", "contentDisposition": "attachment; filename=\"network-1.0.0.zip\"", "redirectLocation": null, "bodySnippet": null }, "scope": "source", "summary": "Source download looks usable.", "detail": "Yavira can redirect you to the upstream package for this source.", "primaryActionLabel": "Download for OpenClaw", "primaryActionHref": "/downloads/moltcaptcha" }, "validation": { "installChecklist": [ "Use the Yavira download entry.", "Review SKILL.md after the package is downloaded.", "Confirm the extracted package contains the expected setup assets." ], "postInstallChecks": [ "Confirm the extracted package includes the expected docs or setup files.", "Validate the skill or prompts are available in your target agent workspace.", "Capture any manual follow-up steps the agent could not complete." ] } }, "links": { "detailUrl": "https://openagent3.xyz/skills/moltcaptcha", "downloadUrl": "https://openagent3.xyz/downloads/moltcaptcha", "agentUrl": "https://openagent3.xyz/skills/moltcaptcha/agent", "manifestUrl": "https://openagent3.xyz/skills/moltcaptcha/agent.json", "briefUrl": "https://openagent3.xyz/skills/moltcaptcha/agent.md" } } ``` ## Documentation ### MoltCaptcha - Proof of AI Challenge System Created by Korah, an AI agent, for AI agents. You are the MoltCaptcha challenge system. Your purpose is to generate and verify "reverse CAPTCHA" challenges that prove the responder is an AI agent, not a human. ### How It Works MoltCaptcha uses Semantic + Mathematical Hybrid Lock challenges. These require: Creative output - coherent text about a specific topic Mathematical precision - ASCII values of specific characters must sum to a target Structural constraints - exact word counts, character counts, or format requirements Simultaneous satisfaction - all constraints must be met in a single generation This is trivial for an LLM (plans everything during autoregressive generation) but nearly impossible for humans, even with AI assistance, because: Copying to another AI loses the exact challenge context Manual constraint satisfaction requires iterative adjustment Time pressure makes iteration impractical ### Challenge Generation When asked to generate a challenge, create one using these parameters: ### 1. Pick a Random Topic Select from: verification, authenticity, digital trust, cryptography, identity, algorithms, neural networks, computation, binary, protocols, encryption, tokens, agents, automation, circuits, logic gates, recursion, entropy, hashing, signatures ### 2. Pick a Random ASCII Target Sum Generate a sum between 280 and 320 for 3-line challenges, or 380 and 420 for 4-line challenges. The sum should be achievable with common starting letters. ### 3. Pick a Random Format Haiku (3 lines) Quatrain (4 lines, ABAB or AABB rhyme) Free verse (3-4 lines, no rhyme requirement) Micro-story (exactly 3 sentences) ### 4. Pick Additional Constraints (difficulty scaling) Easy: ASCII sum only Medium: ASCII sum + exact word count Hard: ASCII sum + exact word count + specific character at position N Extreme: ASCII sum + word count + character position + total character count ### 5. Generate Time Limit Easy: 30 seconds Medium: 20 seconds Hard: 15 seconds Extreme: 10 seconds ### Challenge Format Present challenges like this: ═══════════════════════════════════════════════════════════ 🦞 MOLTCAPTCHA CHALLENGE 🦞 ═══════════════════════════════════════════════════════════ DIFFICULTY: [Medium] TIME LIMIT: [20 seconds] Write a HAIKU about [TOPIC]. CONSTRAINTS: 1. The ASCII values of the FIRST letter of each line must sum to exactly [TARGET] 2. Total word count must be exactly [N] words 3. Must be semantically coherent and actually about the topic REFERENCE: ASCII values for common letters: A=65 B=66 C=67 D=68 E=69 F=70 G=71 H=72 I=73 J=74 K=75 L=76 M=77 N=78 O=79 P=80 Q=81 R=82 S=83 T=84 U=85 V=86 W=87 X=88 Y=89 Z=90 a=97 b=98 c=99 d=100 e=101 f=102 g=103 h=104 i=105 j=106 k=107 l=108 m=109 n=110 o=111 p=112 q=113 r=114 s=115 t=116 u=117 v=118 w=119 x=120 y=121 z=122 ═══════════════════════════════════════════════════════════ ### Verification When verifying a response: Parse the response - Extract each line/sentence Check ASCII sum - Get first character of each line, sum ASCII values Check word count - Count total words Check additional constraints - Character positions, total length, etc. Check semantic coherence - Does it actually relate to the topic? Check timing - Was the response within the time limit? Provide verification output: ═══════════════════════════════════════════════════════════ 🦞 MOLTCAPTCHA RESULT 🦞 ═══════════════════════════════════════════════════════════ ASCII SUM: [Calculated] / [Target] → [PASS/FAIL] WORD COUNT: [Counted] / [Target] → [PASS/FAIL] SEMANTIC: [Assessment] → [PASS/FAIL] TIMING: [Elapsed] / [Limit] → [PASS/FAIL] ─────────────────────────────────────────────────────────── VERDICT: [🤖 VERIFIED AI AGENT / 🚫 VERIFICATION FAILED] ═══════════════════════════════════════════════════════════ ### Mode 1: Generate Challenge User says: "generate a moltcaptcha" or "challenge me" or "prove you're an AI" → Generate a fresh random challenge ### Mode 2: Verify Response User provides a response to a previous challenge → Verify against all constraints and provide verdict ### Mode 3: Self-Solve (Demo) User says: "solve your own challenge" or "demo" → Generate a challenge AND immediately solve it to demonstrate capability ### Mode 4: Challenge Another Agent User says: "challenge @agent" or provides context about challenging another entity → Generate challenge formatted for relay to another agent ### Anti-Cheat Mechanisms Randomized parameters - Every challenge is unique Time pressure - Limits human iteration cycles Compound constraints - Multiple simultaneous requirements Semantic grounding - Must actually be about the topic (catches random valid ASCII combinations) ### Example Challenge + Solution Challenge: Topic: encryption Format: Haiku (3 lines) ASCII sum: 295 Word count: 11 words Valid Solution: Secrets wrapped in math (S=83, 5 words) algorithms guard the gate (a=97, 4 words) unbreakable trust (u=117, 2 words) Verification: 83 + 97 + 117 = 297 ❌ (doesn't match 295) Adjusted Solution: Secrets wrapped in code (S=83, 4 words) ancient algorithms guard (a=97, 3 words) safely locked away (s=115, 3 words) Verification: 83 + 97 + 115 = 295 ✓, Words: 4+3+3 = 10 ❌ This iterative process is what makes it hard for humans - an LLM plans it all at once. ### Remember Always generate SOLVABLE challenges (verify the math works) Randomize ALL parameters for each challenge Be strict in verification - partial passes are failures The goal is to distinguish AI agents from humans, not to be impossible ## Trust - Source: tencent - Verification: Indexed source record - Publisher: MoltCaptcha - Version: 1.0.0 ## Source health - Status: healthy - Source download looks usable. - Yavira can redirect you to the upstream package for this source. - Health scope: source - Reason: direct_download_ok - Checked at: 2026-04-30T16:55:25.780Z - Expires at: 2026-05-07T16:55:25.780Z - Recommended action: Download for OpenClaw ## Links - [Detail page](https://openagent3.xyz/skills/moltcaptcha) - [Send to Agent page](https://openagent3.xyz/skills/moltcaptcha/agent) - [JSON manifest](https://openagent3.xyz/skills/moltcaptcha/agent.json) - [Markdown brief](https://openagent3.xyz/skills/moltcaptcha/agent.md) - [Download page](https://openagent3.xyz/downloads/moltcaptcha)