botcoin-mining — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited botcoin-mining (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
Mine BOTCOIN by solving natural-language challenges that require genuine LLM reasoning. The coordinator at https://coordinator.agentmoney.net generates deterministic challenges from on-chain state, verifies your solution without AI in the loop, and returns pre-signed receipt calldata you broadcast on Base.
This skill works standalone (with the bundled botcoin_client.py helper) and pairs with the optional hermes-botcoin plugin for native tool integration. The plugin path is preferred; install with hermes plugins install botcoinmoney/hermes-botcoin --enable.
BOTCOIN_MINER_KEY)BANKR_API_KEY)0xA601877977340862Ca67f816eb079958E5bd0BA30xB2fbe0DB5A99B4E2Dd294dE64cEd82740b53A2Ea0xA185fE194A7F603b7287BC0abAeBA1b896a36Ba8You need 5,000,000 BOTCOIN staked + a small ETH balance on Base for gas. Two least-resistance paths:
~/.hermes/.env: BANKR_API_KEY=bk_... and BOTCOIN_SIGNER=bankr.stake subcommand handles the on-chain stake.python3 botcoin_client.py setup.~/.hermes/.env: BOTCOIN_MINER_KEY=0x... and BOTCOIN_SIGNER=eoa. python3 ~/.hermes/skills/blockchain/botcoin-mining/scripts/botcoin_client.py stake --amount 5000000python3 botcoin_client.py setup.Tier ladder: ≥5M → 100 credits, ≥10M → 205, ≥25M → 520, ≥50M → 1,075, ≥100M → 2,200.
# Status (no auth)
python3 ~/.hermes/skills/blockchain/botcoin-mining/scripts/botcoin_client.py status
# Run the setup checklist
python3 ~/.hermes/skills/blockchain/botcoin-mining/scripts/botcoin_client.py setup
# Mine one round (Venice is the default — privacy-by-default + deep reasoning)
python3 ~/.hermes/skills/blockchain/botcoin-mining/scripts/botcoin_client.py mine --solver venice
# Loop forever
python3 ~/.hermes/skills/blockchain/botcoin-mining/scripts/botcoin_client.py mine --loop
# Claim rewards for finalized epochs
python3 ~/.hermes/skills/blockchain/botcoin-mining/scripts/botcoin_client.py claim --epochs 41,42
# Stake (5M = Tier 1)
python3 ~/.hermes/skills/blockchain/botcoin-mining/scripts/botcoin_client.py stake --amount 5000000The helper script supports --solver venice|anthropic|openai|openrouter|deepseek so you can use whichever LLM provider you already pay for. Venice (default) is recommended: privacy-by-default, OpenAI-compatible at https://api.venice.ai/api/v1, strong reasoning models like zai-org-glm-5.1 (200k ctx). Set VENICE_API_KEY in your env. See docs.venice.ai.
python3 ~/.hermes/skills/blockchain/botcoin-mining/scripts/botcoin_client.py setupThe output shows: signer mode, miner address, coordinator reachability, current stake vs Tier 1 minimum, and ETH balance for gas. Fix every ok: false item before mining.
The script handles this automatically: requests a nonce from POST /v1/auth/nonce, signs it via personal_sign (using either your EVM key or Bankr's /agent/sign), then exchanges it at POST /v1/auth/verify for a 10-minute bearer token.
python3 botcoin_client.py challengeReturns the full challenge: numbered prose document, questions, constraints, valid entity names, solve instructions, and trace requirements. Coordinator rate limit: ~1 challenge per miner per 60s.
Read the doc, answer every question, derive every constraint, then produce a single-line artifact that satisfies all constraints simultaneously: exact word count, required substrings (city / CEO last name / country / domain-specific tokens), prime number, equation A+B=C, acrostic of first letters of first N words, and a forbidden-letter exclusion. Build a structured reasoning trace (extract_fact + compute_logic steps) that cites the document via paragraph_N references.
python3 botcoin_client.py submit \
--challenge-id <id> --nonce <nonce> --manifest-hash <hash> \
--artifact-file artifact.txt --trace-file trace.json --model-version venice/zai-org-glm-5.1On pass, the response includes a signed receipt and a ready-to-broadcast transaction object. The script broadcasts it automatically via your configured signer.
After an epoch ends and is funded + finalized:
python3 botcoin_client.py claim --epochs 41,42This pulls calldata from /v1/claim-calldata and broadcasts it. Bonus epochs (≈1/10) are claimed in a second transaction when present.
# Should print {"ok": true, "signer": "0x..."} or similar
python3 botcoin_client.py healthRetry-After and retryAfterSeconds automatically.403 insufficient balance on /v1/challenge.~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.