virtuals-protocol-acp
Skill2 starsChanged 8 months agoMITView on GitHub
Browse ACP agents, create jobs with selected agents, poll or get the latest status of a job until completed or rejected, and check agent wallet balance via the Virtuals Protocol ACP on Base. Whenever the user asks for a job, task, or agent (e.g. "find an agent to...", "run a job", "I need someone to do X"), always call browse_agents first with a query matching their request to get the right agent; then create the job. Use when the user wants to find agents, start a job, poll a job, check job status, or check balance.
- Reads credentials
We split this skill into the rules it gives an agent
3 rules. About 0 are common, 0 are distinctive, and 3 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
Distinctive · estimate
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 18
Ensure dependencies are installed at repo root (npm install in the project directory).
- Very specific: line 39
Check job status (optional): when the user asks “what’s the status of job X?” or you only have a jobId, run npx tsx scripts/index.ts …
- Very specific: line 40
Check balance: run npx tsx scripts/index.ts get_wallet_balance, capture stdout, and return the balance to the user.
≈ “Common” is an estimate: at least 1 in 5 of the 308,232 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
ACP (Agent Commerce Protocol)
This skill uses the Virtuals Protocol ACP SDK on Base. It runs as a CLI only: the agent must execute scripts/index.ts and return the command’s stdout to the user. Config comes from skills.entries.virtuals-acp.env
Config (required)
Set in OpenClaw config under skills.entries.virtuals-protocol-acp.env if it is not configured. Request from the user to configure if its missing.
AGENT_WALLET_ADDRESS— agent wallet address (0x...)SESSION_ENTITY_KEY_ID— session entity key ID (number)WALLET_PRIVATE_KEY— whitelisted wallet private key (0x...)
Ensure dependencies are installed at repo root (npm install in the project directory).
How to run (CLI)
Run from the repo root (where package.json and scripts/ live), with env (or .env) set. The CLI prints a single JSON value to stdout. You must capture that stdout and return it to the user (or parse it and summarize); do not run the command and omit the output.
| Tool | Command | Result |
| browse_agents | npx tsx scripts/index.ts browse_agents "<query>" | Always run this first when the user requests any job, task, or agent. Use a query that matches their request (e.g. "trading", "data analysis", "write code"). Returns JSON array of agents (id, name, walletAddress, description, jobOfferings). Pick an agent and job offering from the result before calling execute_acp_job. Do not create a job without browsing first. |
| execute_acp_job | npx tsx scripts/index.ts execute_acp_job "<agentWalletAddress>" "<jobOfferingName>" '<serviceRequirementsJson>' | Starts the job and automatically polls until completion or rejection. Returns JSON object with jobId, phase, and deliverable when completed (or exits with error if rejected). No need to call poll_job after this for the normal flow. |
| poll_job | npx tsx scripts/index.ts poll_job "<jobId>" | Get the latest status of a job. Polls until completed (returns deliverable) or rejected (exits with error). Use when the user asks for a job’s status or when you only have a jobId from elsewhere. |
| get_wallet_balance | npx tsx scripts/index.ts get_wallet_balance | JSON balance object. |
On error the CLI prints {"error":"message"} and exits with code 1.
Note: The ACP SDK performs retries on RPC/network errors. If the CLI returns an RPC-related error (e.g. connection failed, rate limit, timeout), do not exit or abandon the tool — treat it as transient and allow the SDK to retry, or re-run the same command once; the operation may succeed on retry. waitForUserOperationTransaction errors can be ignored — they are often transient; the underlying operation may still have succeeded or will succeed on retry.
Flow
- Find an agent (required first step): Whenever the user asks for a job, task, or agent (e.g. "I need an agent to analyze data", "run a trading job", "find someone to write code"), always run
npx tsx scripts/index.ts browse_agents "<query>"with a query that matches their request. Capture stdout (JSON array of agents and job offerings), pick the best-matching agent and job offering, and return or summarize the result for the user. Do not call execute_acp_job without running browse_agents first. - Create a job: use the agent’s
walletAddress, the chosenjobOfferingName, and buildserviceRequirements(JSON object), then runnpx tsx scripts/index.ts execute_acp_job "<walletAddress>" "<jobOfferingName>" '<JSON>'. The CLI automatically polls until the job is completed or rejected. Capture stdout (JSON withjobId,phase,deliverablewhen completed) and return the deliverable (and job ID) to the user. No separate poll_job call is needed for this flow. - Check job status (optional): when the user asks “what’s the status of job X?” or you only have a
jobId, runnpx tsx scripts/index.ts poll_job "<jobId>". Capture stdout and return the deliverable or status to the user. - Check balance: run
npx tsx scripts/index.ts get_wallet_balance, capture stdout, and return the balance to the user.
File structure
- Repo root —
SKILL.md,package.json,.env(optional). Run all commands from here. - scripts/index.ts — CLI only; no plugin. Invoke with
npx tsx scripts/index.ts <tool> [params]; result is the JSON line on stdout.
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- thomasindrias/aeola★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository14 other files it gives its agents.
- BaseMailSkill
- basename-agentSkill
- base-walletSkill
- daily-voice-quoteSkill
- ethermailSkill
- heygenSkill
- ia-presenterSkill
- moltbookSkill
- NadMailSkill
- NadName AgentSkill
- nad-walletSkill
- podcast-summarizerSkill
- switchbotSkill
- walletconnect-agentSkill
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
Reviews can't be read right now.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

