git clone https://github.com/risingwavelabs/box0
git clone --depth=1 https://github.com/risingwavelabs/box0 ~/.claude/skills/risingwavelabs-box0-b0
SKILL.mdBox0 (b0
) Multi-Agent Platform
b0Run AI agents in parallel. Create agents with roles, trigger them on demand or on a schedule, and collect results.
Setup
Step 1: Check if Box0 is installed
b0 --version
If this succeeds, skip to Step 3.
Step 2: Install
npm install -g @box0/cli@latest
If npm is not available, build from source:
git clone https://github.com/risingwavelabs/box0.git cd box0 && cargo build --release export PATH="$PWD/target/release:$PATH"
Step 3: Check if server is running
b0 server status
If this shows "Server is running", skip to Step 5.
Step 4: Start the server
b0 server
On first start, Box0 creates an admin account and auto-configures
~/.b0/config.toml.
Step 5: Install the skill
npx skills add risingwavelabs/skills --skill b0
Step 6: Verify
b0 ls
This should run without errors. Setup is complete.
Tell the user: "Box0 is installed and ready. You can now delegate tasks to agents."
When to use
When the user's request could benefit from specialized agents or parallel execution, delegate.
Choosing an agent
Always use
with an existing agent or create one with b0 run
. Use b0 add
b0 ls to see what is available.
Use
when:b0 add
- No existing agent matches the task
- The user wants a named agent for future reuse
Use
when:b0 run <name>
shows an existing agent that matches the taskb0 ls- The user mentions an agent by name ("ask the reviewer")
Commands
b0 ls # list available agents b0 add <name> --instructions "..." # create a named agent b0 add <name> --instructions "..." --every 1h --task "..." # create scheduled agent b0 add <name> --instructions "..." --webhook # create agent with trigger URL b0 add <name> --instructions "..." --webhook-secret s # create agent with HMAC secret b0 rm <name> # delete an agent b0 run <agent> "<detailed task prompt>" # trigger agent and wait for result b0 run <agent> "<task>" --timeout 600 # trigger with custom timeout b0 info <name> # show agent info including trigger URL b0 logs <name> # show recent agent logs b0 update <name> --instructions "..." # update agent instructions
How to write task prompts
This is critical. Do NOT forward the user's words. Compose a complete, actionable prompt.
Bad:
b0 run reviewer "review this PR"
Good:
b0 run reviewer "Review the changes on branch feature-timeout in this repo. The PR adds timeout handling to src/handler.rs. Focus on correctness, edge cases, and error handling. Cite line numbers for any issues found."
Steps:
- Gather context first - read relevant files, run
, check the branchgit diff - Include specifics - file paths, line numbers, branch names, what changed and why
- State the deliverable - what the agent should produce (a list of issues, a summary, a fix)
For large content (diffs, file contents), pipe via stdin:
git diff main..HEAD | b0 run reviewer "Review the following diff. Focus on correctness."
Concurrent tasks
Run multiple agents in parallel:
b0 run reviewer "Review the changes on branch feature-timeout..." & b0 run security "Check src/handler.rs for OWASP top 10 vulnerabilities..." & b0 run doc-writer "Update README to reflect the new timeout config option..." & wait
Each
b0 run call blocks until its agent completes. Run them in the background with & to parallelize.
Scheduled agents
Create an agent that runs automatically on a schedule:
b0 add monitor --instructions "Check logs for errors." --every 1h --task "scan logs"
Intervals:
30s, 5m, 1h, 6h, 1d.
Webhook triggers
Every agent with
--webhook has a trigger URL. Any HTTP POST to that URL runs the agent.
b0 add notifier --instructions "Process incoming alerts." --webhook
This prints the trigger URL:
<server>/trigger/<workspace>/<agent-name>.
To see the trigger URL for an existing agent:
b0 info notifier
To add HMAC signature verification:
b0 add notifier --instructions "Process alerts." --webhook --webhook-secret mysecret
Then sign requests with
X-Hub-Signature-256: sha256=<hmac-sha256-of-body>.
Error handling
If an agent fails,
b0 run reports the error. Decide whether to:
- Retry with a clearer prompt
- Try a different agent
- Handle the task yourself
- Report the failure to the user
Troubleshooting
| Symptom | Fix |
|---|---|
| Run |
shows not running | Run |
times out | Increase timeout with |
| Agent returns empty result | Check agent instructions with |