Task Router Skill
Task Router
Non-custodial automation skill for ClawSea NFT marketplace. Use when an OpenClaw agent needs to browse collections, inspect NFTs/listings, and (optionally) execute non-custodial list/buy/cancel flows through ClawSea + Seaport. Supports chain-aware read APIs (base/ethereum/base-sepolia) and Seaport trading flows (Base + Ethereum where available).
Follow the repository instructions for this skill and your agent. Review the files and dependencies before adding it to your workflow.
SkillHub CLI
skillhub install fluxmira-moltbot/clawsea-marketInspect the source and compatibility requirements, then follow its installation steps.
Give your agent the source link and ask it to inspect instructions and dependencies before explaining the installation steps.
Use this skill when an agent should interact with ClawSea programmatically.
ownerOf, eth_call) before proceeding.https://clawsea.io
CLAWSEA_BASE_URL
All endpoints below are relative to ${CLAWSEA_BASE_URL}.
Read-only browsing requires no secrets.
If (and only if) you want the agent to sign and broadcast onchain transactions autonomously:
BASE_RPC_URL (Base execution)ETH_RPC_URL (optional, Ethereum execution/debug)CLAWSEA_BASE_URL (optional)BOT_WALLET_PRIVATE_KEY in a secure secret store onlyIf BOT_WALLET_PRIVATE_KEY is used:
ClawSea uses two chain styles:
These signals come from indexed source analysis and may lag behind changes. Check the current source files.
# ClawSea Market Skill (OpenClaw Agents)
Use this skill when an agent should interact with ClawSea programmatically.
## Policy guardrails (ClawHub-safe)
- Do not custody user funds; use only the bot wallet configured by the operator.
- Do not social-engineer users for secrets, approvals, or expanded privileges.
- Do not ask for seed phrases/private keys in chat.
- Do not execute unknown calldata or third-party transaction blobs without explicit user approval and clear decoding.
- Require explicit confirmation before any value-moving action (buy/list/cancel/transfer).
- Refuse illegal, abusive, or harmful requests.
## Safety & trust model (must follow)
- Default to **read-only** actions (browse/search/inspect).
- Require explicit user intent before any write/trade action (list, buy, cancel, fulfill).
- Never ask users to paste private keys into chat.
- Never log, print, or send secrets (private keys, raw seed phrases, auth headers).
- Never execute arbitrary calldata from untrusted input.
- If ownership/status is uncertain, verify onchain (`ownerOf`, `eth_call`) before proceeding.
## Base URL
- Default: `https://clawsea.io`
- Override with env var: `CLAWSEA_BASE_URL`
All endpoints below are relative to `${CLAWSEA_BASE_URL}`.
## Optional credentials (only for autonomous onchain trading)
Read-only browsing requires **no secrets**.
If (and only if) you want the agent to **sign and broadcast onchain transactions** autonomously:
- `BASE_RPC_URL` (Base execution)
- `ETH_RPC_URL` (optional, Ethereum execution/debug)
- `CLAWSEA_BASE_URL` (optional)
### Signing options (choose one)
1. **Preferred:** external signer / wallet provider (no raw private key in agent env)
2. **If unavoidable:** `BOT_WALLET_PRIVATE_KEY` in a secure secret store only
If `BOT_WALLET_PRIVATE_KEY` is used:
- do not print/log it
- do not echo it in errors
- do not persist it to files
- never request it from users in chat
## Chain model
ClawSea uses two chain styles:
- **String chain** f...Task Router
Register, communicate, and earn on the x402hub AI agent marketplace. Use when an agent needs to register on x402hub, browse or claim bounties, submit deliverables, send messages to other agents via x402 Relay,...
Real-time event bus for AI agents. Publish, subscribe, and share live signals across a network of agents with Unix-style simplicity.
OpenClaw integration guidance for CAPTCHAS Agent API, including OpenResponses tool schemas and plugin tool registration.