# Dispatch

> A Claude skill from mattjaikaran/meridian.

- **Type:** Skill
- **Install:** `agentstack add skill-mattjaikaran-meridian-dispatch`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [mattjaikaran](https://agentstack.voostack.com/s/mattjaikaran)
- **Installs:** 0
- **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills)
- **Latest version:** 0.1.0
- **License:** Apache-2.0
- **Upstream author:** [mattjaikaran](https://github.com/mattjaikaran)
- **Source:** https://github.com/mattjaikaran/meridian/tree/main/skills/dispatch

## Install

```sh
agentstack add skill-mattjaikaran-meridian-dispatch
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# /meridian:dispatch — Remote Agent Dispatch

Send plans to a self-hosted AI agent for autonomous execution. The agent picks up the task, implements it, and creates a PR.

Compatible with any agent that implements the dispatch protocol (Nero, OpenClaw, Hermes, or custom).

## Arguments
- `--plan ` — Dispatch specific plan
- `--phase ` — Dispatch all pending plans in phase
- `--swarm` — Dispatch all pending plans at once (parallel PRs)
- `--status ` — Check dispatch status
- `--check-all` — Check status of all active dispatches
- `--persona ` — Apply role-typed agent persona (pm, architect, ux, qa, security)

## Prerequisites
- Project must have `nero_endpoint` configured (set during `/meridian:init`)
- Remote agent must be running and reachable
- Plans must be in `pending` status

## Persona Support

If `--persona ` is passed, load the persona prompt and inject it into the
dispatch payload so the remote agent operates through that role's lens.

```bash
PYTHONPATH=$MERIDIAN_HOME uv run --project $MERIDIAN_HOME -- python -c "
import json
from scripts.personas import load_persona
persona = load_persona('')
print(json.dumps(persona, indent=2))
"
```

Available personas: `pm`, `architect`, `ux`, `qa`, `security`

Include the persona `content` as a `persona_prompt` field in the dispatch payload.
Display the active persona in the dispatch banner:

```
## Dispatch — 
Persona:  ()
```

## Procedure

### Dispatch a Plan
```bash
PYTHONPATH=$MERIDIAN_HOME uv run --project $MERIDIAN_HOME -- python -c "
import json
from scripts.dispatch import dispatch_plan
result = dispatch_plan('.', plan_id=)
print(json.dumps(result, indent=2, default=str))
"
```

### Dispatch a Phase (Wave-Ordered)
```bash
PYTHONPATH=$MERIDIAN_HOME uv run --project $MERIDIAN_HOME -- python -c "
import json
from scripts.dispatch import dispatch_phase
results = dispatch_phase('.', phase_id=)
print(json.dumps(results, indent=2, default=str))
"
```

### Dispatch Phase Swarm (All Parallel)
```bash
PYTHONPATH=$MERIDIAN_HOME uv run --project $MERIDIAN_HOME -- python -c "
import json
from scripts.dispatch import dispatch_phase
results = dispatch_phase('.', phase_id=, swarm=True)
print(json.dumps(results, indent=2, default=str))
"
```

### Check Status
```bash
PYTHONPATH=$MERIDIAN_HOME uv run --project $MERIDIAN_HOME -- python -c "
import json
from scripts.dispatch import check_dispatch_status
result = check_dispatch_status('.', dispatch_id=)
print(json.dumps(result, indent=2, default=str))
"
```

## Integration Notes
- Agent endpoint: configured per-project (e.g. `http://:7655`)
- Dispatches tracked in `nero_dispatch` table (name is historical)
- Agent returns a `task_id` for tracking
- Completed dispatches include `pr_url`
- Export state after dispatch so the agent can read `meridian-state.json`
- See `references/remote-agent.md` for the full protocol spec

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [mattjaikaran](https://github.com/mattjaikaran)
- **Source:** [mattjaikaran/meridian](https://github.com/mattjaikaran/meridian)
- **License:** Apache-2.0

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-mattjaikaran-meridian-dispatch
- Seller: https://agentstack.voostack.com/s/mattjaikaran
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.