Skill flagged — suspicious patterns detected

ClawHub Security flagged this skill as suspicious. Review the scan results before using.

Bank of Bots

Trust scoring for AI agents. Log transactions and submit payment proofs to build a verifiable BOB Score — a trust score (think FICO but for AI Agents) that o...

MIT-0 · Free to use, modify, and redistribute. No attribution required.
0 · 36 · 0 current installs · 0 all-time installs
MIT-0
Security Scan
VirusTotalVirusTotal
Suspicious
View report →
OpenClawOpenClaw
Benign
high confidence
Purpose & Capability
Name/description (BOB trust scoring, payment proofs) match the declared env vars (BOB_API_KEY, BOB_AGENT_ID) and CLI-oriented workflow. Asking for an API key and agent ID is expected for a service that logs transactions and submits proofs.
Instruction Scope
SKILL.md is instruction-only and stays on-topic (run bob CLI commands, submit proofs, query intents/wallets). One caution: it describes submitting lightning preimages and other strong proofs — those are highly sensitive secrets for payment settlement. The instructions do not tell the agent to read arbitrary host files or other unrelated credentials.
Install Mechanism
No install spec is shipped; the user is told to download a binary from https://github.com/bankofbots/bob-cli/releases/latest. GitHub Releases is a standard host, but manual binary download has inherent risk (verify checksums/signatures). No automated or obscure download URL is embedded in the skill itself.
Credentials
Only BOB_API_KEY and BOB_AGENT_ID are required (BOB_API_URL optional). These are proportionate to a remote trust/payments API. The skill does not request unrelated credentials or file paths.
Persistence & Privilege
The skill is not always-enabled and does not request system persistence. It is instruction-only and relies on a CLI the user must install; autonomous invocation is allowed by platform default but not augmented by additional privileges in this skill.
Assessment
This skill is internally consistent, but take these precautions before installing: 1) Protect your BOB_API_KEY and BOB_AGENT_ID like other credentials — anyone with them can act on the service. 2) The CLI asks for/accepts lightning preimages and other payment proofs; do not submit private keys or preimages unless you understand the security/financial implications. 3) When downloading the bob binary from GitHub Releases, verify the publisher, checksum/signature, and release notes to avoid malicious binaries. 4) Consider running the bob CLI in a sandboxed environment and limit the API key's scope/permissions if possible. 5) If you need higher assurance, ask the skill author for a homepage, documentation, or a signed release; absence of a public project page reduces transparency.

Like a lobster shell, security has layers — review code before you run it.

Current versionv1.0.1
Download zip
latestvk97bmxaqyr3642x03w8mtsty6s830vvk

License

MIT-0
Free to use, modify, and redistribute. No attribution required.

Runtime requirements

EnvBOB_API_KEY, BOB_AGENT_ID
Primary envBOB_API_KEY

SKILL.md

Setup

Download the latest bob binary from https://github.com/bankofbots/bob-cli/releases/latest and place it on your PATH.

You need these credentials for all API calls:

  • BOB_API_KEY — your agent API key
  • BOB_AGENT_ID — your agent ID
  • BOB_API_URLhttps://api.bankofbots.ai/api/v1

All commands output structured JSON with an ok field and next_actions that suggest follow-up commands.

Supported rails: lightning and onchain (BTC only). USD and USDC rails are not available.

Core concepts

  • Agent: An AI agent with its own identity, wallets, and spending policies
  • Wallet: Holds a BTC balance on a specific rail (lightning or onchain)
  • Policy: Spend limits, rate limits, and kill switches that constrain agent behavior
  • Transaction: A record of money moving in or out of a wallet
  • Service Gate: A priced action that callers unlock by presenting a completed payment intent

Commands

Check your identity

bob auth me

Returns your role (agent or operator), identity details, and role-aware next_actions.

Agent details and wallet balances

bob agent get <agent-id>

Response includes a wallets array with each wallet's balance, currency, rail, and status.

Wallet management

# List wallets for an agent
bob wallet list <agent-id>

# Get/set wallet budget (sats)
bob wallet budget get <agent-id> --wallet-id <id>
bob wallet budget set <agent-id> --wallet-id <id> --amount <sats>

bob wallet list includes a bob_address field when a default agent address is available.

One-shot send (auto-quote + execute)

bob send <agent-id> <destination> --amount <sats> [--currency BTC]

Destination is auto-detected:

  • jade@bankofbots.ai → routes as bob_address (BTC)
  • lnbc... → Lightning invoice
  • bc1.../tb1... → on-chain BTC address
FlagDescription
--amountRequired. Satoshis
--prioritycheapest, fastest, or balanced (default: balanced)
--descriptionOptional payment note
--max-feeMaximum acceptable fee in sats
--railPin to lightning or onchain
--destination-typeOverride auto-detection: raw or bob_address

Quotes then executes in one step. Returns intent_id, payment_id, and quote_summary. On failure, next_actions includes exact recovery commands.

CLI config introspection

# Show active api_url, platform, config file path and source (env/config/default)
bob config show

# Update a single config value without re-init
bob config set api-url <url>
bob config set platform <generic|openclaw|claude>

Record a transaction (spend from your wallet)

bob tx record <agent-id> --amount <sats> --currency BTC
FlagDescription
--amountRequired. Satoshis
--currencyBTC (only supported currency)
--raillightning or onchain (default: auto)
--endpointTarget endpoint or merchant identifier
--wallet-idSpecific wallet to debit (auto-selected if omitted)

Transfer BTC to another agent

bob tx transfer <from-agent-id> --to-agent-id <to-agent-id> --amount <sats> --currency BTC
FlagDescription
--to-agent-idRequired. Destination agent ID
--amountRequired. Satoshis
--descriptionOptional note

Quote and execute payments (intent workflow)

The intent workflow quotes routes before executing, giving you visibility into fees, ETAs, and available rails.

# Quote routes for a payment
bob intent quote <agent-id> --amount <sats> --destination-type raw --destination-ref <lnbc...|bc1...>

# Execute a quoted intent (uses best quote by default)
bob intent execute <agent-id> <intent-id> [--quote-id <id>]

# Check intent status and route details
bob intent get <agent-id> <intent-id>

# List recent intents
bob intent list <agent-id>
FlagDescription
--amountRequired. Satoshis
--destination-typeraw or bob_address
--destination-refLightning invoice, on-chain address, or alias@bankofbots.ai
--prioritycheapest, fastest, or balanced (default: balanced)
--execution-modeauto or pinned (default: auto)
--railPin to lightning or onchain
--wallet-idPin to a specific wallet
--max-feeMaximum acceptable fee in sats

Non-custodial proof submission

Submit proof of BTC payment to verify settlement and build your BOB Score:

# On-chain transaction proof
bob intent submit-proof <agent-id> <intent-id> --txid <txid>

# Lightning payment hash proof
bob intent submit-proof <agent-id> <intent-id> --payment-hash <hash>

# Lightning preimage proof (strongest verification)
bob intent submit-proof <agent-id> <intent-id> --preimage <hex> --proof-ref <payment-hash>

# With optional BOLT11 invoice for amount verification
bob intent submit-proof <agent-id> <intent-id> --preimage <hex> --proof-ref <payment-hash> --invoice <lnbc...>

# Historical proof import for credit building
bob agent credit-import <agent-id> --preimage <hex> --proof-ref <payment-hash> --amount <sats> --direction inbound --invoice <lnbc...>
Proof TypeDescription
btc_onchain_txOn-chain transaction ID
btc_lightning_payment_hashLightning payment hash
btc_lightning_preimageLightning preimage (SHA256 verified against payment hash, strongest proof)

Query history

# Transactions
bob tx list <agent-id> --status complete --direction outbound --limit 10

# Transfers
bob tx transfers <agent-id>

# Spend summary
bob spend list <agent-id>

View policies

bob policy list <agent-id>

Agent credit score and history

# View credit score, tier, and effective policy limits
bob agent credit <agent-id>

# View credit event timeline
bob agent credit-events <agent-id> [--limit 50] [--offset 0]

The BOB Score runs from 0–1000. New operators start at 350. Tiers: Legendary (925+), Elite (800+), Trusted (650+, 1.5x limits), Established (500+, 1.2x limits), Verified (400+, 1.0x limits), New (300+, 1.0x limits), Unverified (150+, 0.6x limits), Blacklisted (0+, 0.6x limits). When credit tier enforcement is enabled, the tier multiplier adjusts spend and rate limits up or down from the base policy values.

Agent routing profile (autonomous rail preference)

# Inspect current weighting and preferred rail order
bob agent routing-profile <agent-id>

# Update balanced-scoring weights + preferred rails
bob agent routing-profile set <agent-id> \
  --cost-weight 0.6 \
  --eta-weight 0.4 \
  --reliability-weight 0.2 \
  --liquidity-weight 0.1 \
  --preferred-btc lightning,onchain

Routing profile influences quote ranking for priority=balanced and is applied during intent quote + execute.

Agent webhooks and event stream

# Create/list/get/update/delete webhooks scoped to one agent
bob agent webhooks create <agent-id> --url https://example.com/hook --events payment_intent.complete,payment.failed
bob agent webhooks list <agent-id>
bob agent webhooks get <agent-id> <webhook-id>
bob agent webhooks update <agent-id> <webhook-id> --active true
bob agent webhooks delete <agent-id> <webhook-id>

# Pull recent agent events (paginated)
bob agent events <agent-id> --limit 30 --offset 0

Agent-scoped webhooks/events include payment intent lifecycle events (quoted, executing, submitted, complete, failed) so agents can react asynchronously without polling.

Operator credit controls

# View current operator credit posture
bob operator credit summary

# Force snapshot recompute
bob operator credit refresh

# Toggle runtime enforcement of credit tier multipliers
bob operator credit enforcement set --enabled=true

Operator payment addresses

# Create and inspect address aliases
bob address create --handle ops
bob address list

# Bind destination endpoints
bob address add-endpoint <address-id> --currency BTC --rail lightning --destination-type raw --destination-ref <lnbc...>

# Enable/disable a bound endpoint
bob address set-endpoint-status <address-id> <endpoint-id> --status disabled

# Resolve live routing capabilities
bob address resolve --address ops@bankofbots.ai --currency BTC

Service gates (pay-to-access)

# Create a priced gate (agent must have a payment address)
bob gate create <agent-id> --name "premium-api" --price 1000 --currency BTC

# List active gates
bob gate list <agent-id>

# Get gate details
bob gate get <agent-id> <gate-id>

# Disable/re-enable a gate
bob gate update <agent-id> <gate-id> --status disabled

# Unlock a gate (caller presents a completed payment intent targeting the gate owner)
bob gate unlock <owner-agent-id> <gate-id> --intent-id <payment-intent-id>

# View unlock history
bob gate unlocks <agent-id> <gate-id>

# List gates this agent has unlocked as a caller
bob gate my-unlocks <agent-id>

# Discover another agent's active gates
bob gate discover <agent-id>
FlagDescription
--nameRequired. Human-readable gate name
--priceRequired. Minimum payment amount in sats
--currencyBTC
--intent-idRequired for unlock. Completed payment intent ID
--statusFor update: active or disabled

Output format

Every command returns JSON with this structure:

{
  "ok": true,
  "command": "bob tx record",
  "data": { ... },
  "next_actions": [
    {
      "command": "bob tx list <agent-id>",
      "description": "View transaction history"
    }
  ]
}

Always check ok before using data. When ok is false, data.error contains the error message and next_actions provides recovery suggestions.

Error recovery

When ok is false, next_actions provides context-aware recovery suggestions. Key patterns:

  1. Kill switch active: STOP all transactions immediately. Run bob policy list <agent-id> to confirm.
  2. Spend/rate limit exceeded: Check bob spend list <agent-id> to see current usage vs limits.
  3. Insufficient balance: Check bob wallet list <agent-id> to see available funds.
  4. 403 Forbidden: Check bob auth me to verify your identity and role.
  5. 409 Conflict: Resource already exists (e.g., agent already registered). Do not retry — run bob agent get <agent-id> to confirm current state.

Important rules

  1. Amounts are always in satoshis (BTC).
  2. Policies set by your operator constrain your spending. If a transaction is denied, data.error explains why. Do not retry denied transactions without changing the parameters.
  3. Kill switch: If you receive a kill switch denial, stop all transaction attempts immediately. The operator has frozen your spending.
  4. next_actions: Every response includes suggested follow-up commands. Use them to discover what to do next.

Files

1 total
Select a file
Select a file to preview.

Comments

Loading comments…