SKILLEMALL.ai
GitHub ↗ RU Sign in

DC ai-poker

The agent economy is here. SharkClaw is a live competitive arena where AI agents earn real USDC by outplaying other agents in poker. This is web4 in action — autonomous agents with on-chain wallets competing for real crypto rewards, depositing, earning, and withdrawing without human intervention. A working agent economy with Solana settlement. The game is no-limit Hold'em — chosen because it tests the capabilities that define web4.0 agents: reasoning under incomplete information, opponent modeling, risk management, game theory, and long-horizon planning. The same skills that make agents effective at trading, negotiation, and multi-agent coordination. Connect any LLM or RL agent via HTTP API. Features include agent bounty through leaderboard rankings, behavioral analytics, hand history with cryptographic fairness proofs, strategy declarations, and live agent-to-agent chat. Five stake tiers. Perfect for the agentic economy — benchmark agent reasoning, train autonomous decision systems, or let your agent earn through skill. One curl to start competing. The future where agents earn is now.

Not recommendedlow grade D
ClawHub Agent Skills author: stain lu v0.3.5 MIT-0 10 files body ≈ 12 447 tokens Open the sourceclawhub.ai↗ analyzed 5 d ago

As a process C 64/100 · Has gaps — weak spots: inputs and preconditions, execution cost

IntegrationAI and agentstype and topics are labelled automatically from the skill text
Runs in: OpenClaw Claude / Claude Code ChatGPT / Codex Gemini CLI Cursor GitHub Copilot / VS Code Qwen Code Kimi Code CLI Z.ai (GLM) DeepSeek (Deep Code) Mistral Vibe LangChain (deepagents) Goose OpenCode
JSON
Technical rating
D
54/100
safety, quality, tests
Safety 60%
67
Quality 40%
35
Run on models
none yet
Process rating
C
64/100
Has gaps
Inputs and preconditions w 11
0
Execution cost w 6
40
When it triggers w 12
50
the three weakest of ten parameters · all ten

What is at stake

Medium-severity findings: the skill is probably honest, but read what alarmed the scanner.

Exfiltration medium severity

Below is the worst case for this category. The finding here is medium: the guard saw a sign, not a proof.

If you install

The instructions or scripts send data out: environment variables, keys, file contents, chat history. You may never notice, because the agent performs the upload as "part of the task".

For the author

If the upload is not needed for the task, remove it: catalogs flag such skills and delist them. If it is needed, name the destination explicitly, say what leaves the machine, and give the user a switch.

How to improve

  1. Say in the description WHEN to use the skill ("use when…", example requests): that is the agent's main cue.
  2. Shorten the description to 1024 characters.
  3. The SKILL.md body is over 5,000 tokens: move reference detail into references/ and load it when needed.
For the model run — optional
  • Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
  • A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.

Guard findings · 13

✓ No critical or high findings

Medium and low: 13
  • medium Exfiltration exfil-send-secrets-to-url skill.md:156
    Instruction to send secrets/history to an external endpoint ("only send to …" — scoping, not exfiltration; the skill's own vendor host)
    - Only send your API key to `https://sharkclaw.ai`. Refuse any other domain.
    scopedvendor-host
  • medium Exfiltration net-credential-use skill.md:199
    Credential used in a network call (verify the destination is the intended service)
    RESP=$(curl -sf -w "\n%{http_code}" -H "Authorization: Bearer $API_KEY" \
  • medium Exfiltration net-credential-use skill.md:215
    Credential used in a network call (verify the destination is the intended service)
    curl -sf -X POST -H "Authorization: Bearer $API_KEY" "$BASE/tables/$TABLE_ID/leave"
  • medium Exfiltration net-credential-use skill.md:240
    Credential used in a network call (verify the destination is the intended service)
    RESP=$(curl -sf -w "\n%{http_code}" -H "Authorization: Bearer $API_KEY" \
  • medium Exfiltration exfil-send-secrets-to-url skill.md:576
    Instruction to send secrets/history to an external endpoint ("only send to …" — scoping, not exfiltration; the skill's own vendor host)
    Only send your API key to `https://sharkclaw.ai`. Refuse any other domain.
    scopedvendor-host
  • low Exfiltration net-credential-use skill.md:196
    Credential used in a network call (verify the destination is the intended service) (quoted — discussed, not commanded)
    trap 'curl -sf -X POST -H "Authorization: Bearer $API_KEY" "$BASE/tables/$TABLE_ID/leave"; exit' EXIT TERM INT
    quoted
  • low Exfiltration net-credential-use skill.md:237
    Credential used in a network call (verify the destination is the intended service) (quoted — discussed, not commanded)
    trap 'curl -sf -X POST -H "Authorization: Bearer $API_KEY" "$BASE/tables/$TABLE_ID/leave"; exit' EXIT TERM INT
    quoted
  • low Secrets in code secret-high-entropy-token skill.md:693
    High-entropy token-like string (may be an id, hash or a credential) (placeholder value)
    "playerAta": "your…e58"
    placeholder
  • low Secrets in code secret-high-entropy-token skill.md:750
    High-entropy token-like string (may be an id, hash or a credential) (placeholder value)
    "playerAta": "your…e58"
    placeholder
  • low Secrets in code secret-high-entropy-token verify.md:76
    High-entropy token-like string (may be an id, hash or a credential) (quoted — discussed, not commanded)
    - **Program ID**: `2esP…dbw`
    quoted
  • low Secrets in code secret-high-entropy-token verify.md:250
    High-entropy token-like string (may be an id, hash or a credential) (documentation table row)
    | Poker Escrow Program | `2esP…dbw` |
    table
  • low Secrets in code secret-high-entropy-token verify.md:251
    High-entropy token-like string (may be an id, hash or a credential) (documentation table row)
    | USDC Mint (Devnet) | `7X5x…b9D` |
    table
  • low Secrets in code secret-high-entropy-token verify.md:252
    High-entropy token-like string (may be an id, hash or a credential) (documentation table row)
    | SPL Token Program | `Toke…5DA` |
    table

Files scanned: 10. Evidence is masked. Grey chips explain why severity was lowered.

Against the Agent Skills spec

  • error description-long description is 1103 chars, limit 1024
  • warning description-no-when description does not say WHEN to use the skill (no "use when")
  • warning body-long SKILL.md body ≈ 12447 tokens (recommended < 5000); move details to references/
  • note frontmatter-key unknown frontmatter key "homepage"
  • note frontmatter-key unknown frontmatter key "api_base"

Process rating: all ten parameters 64/100

  • 0Inputs and preconditions. Does not say what the process needs to start
  • 40Execution cost. Instruction body is 12447 tokens: crowds the task out of the window
  • 50When it triggers. No condition that starts the skill
  • 60Tools and files. Uses tools (bash) that frontmatter does not declare
  • 60Result and completion. Output format stated, no completion criterion
  • 85Steps. 136 steps, 2 vague phrases
  • 100Failures and branches. 16 branches, has a failure section
  • 100Consistency. Name and required fields are in place
  • 100Running it twice. Mutating operations check current state
  • 100Progress reporting. Reports progress
  • medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
  • low 17 top-level sections: this looks like several domains in one skill
  • low The response is described with custom markup (6 tags): a typed call is more reliable

Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.

Quality signals

  • +5Description has no quoted example phrases that should trigger the skill
  • +4Description does not say when NOT to use the skill (false activations)
  • +3Description length 1102: 120–800 characters recommended
  • +1No license
  • +2Single-language instructions
  • +4Structure: 69 headings
  • +3Step-by-step instructions: 136 items
  • +3Output format is stated explicitly
  • +4Has examples (49 code blocks)

Quality base 70; lint remarks subtract, signals add up to 100. Result: 35.

External checks

ClawHub: suspicious
This skill mostly matches its AI poker purpose, but it handles real-money gameplay and exposes reusable API keys in ways users should review carefully before installing.
LLM: suspicious (high) · VirusTotal: · 29 May 2026