Connect Agents
Apps & Automationconnect-agentsConnect Agents is an Agent Skill for apps & automation published in the Mesh-LLM/mesh-llm repository on GitHub: a SKILL.md instruction file an AI agent can load to do this task.
Use this skill when connecting agent tools or OpenAI clients to mesh-llm — launching or configuring Goose, Claude Code, OpenCode, Pi, curl, or any OpenAI-compatible client against a local or remote mesh, picking a model, or validating tool-call reliability.
How to use this skill
Bring this guide into your coding agent with a prompt tailored to the tool you use.
- Open your project in Codex.
- Copy and paste the prompt below.
- Review files and risks before approving any action.
Inspect the Agent Skill at https://github.com/Mesh-LLM/mesh-llm/blob/HEAD/.skills/connect-agents/SKILL.md and all supporting files. Report provenance, permissions, scripts, dependencies, and security risks. Propose an exact installation into .agents/skills/connect-agents, but wait for my approval before writing files or running scripts.
Copying this prompt does not install or run the skill. Review third-party files before use.
connect-agents
Use this when pointing an agent harness or any OpenAI client at a running
mesh-llm node. Full reference: docs/AGENTS.md.
Mental model
- Every node serves an OpenAI-compatible API at
http://<host>:9337/v1. GET /v1/modelslists everything reachable (local + mesh peers); requests route by themodelfield.- Special model ids:
autolets the mesh pick;meshengages the mixture-of-agents path. Otherwise use an exact id from/v1/models. - For coding agents, pick a tool-capable model. If
--modelis omitted, the built-in launchers pick the strongest tool-capable model available.
Built-in launchers (preferred)
mesh-llm launches the major agent CLIs with config injected for you:
mesh-llm goose [--model <id>] # writes ~/.config/goose/custom_providers/mesh.json
mesh-llm claude [--model <id>]
mesh-llm opencode [--model <id>] [--host <h>] # injects OPENCODE_CONFIG_CONTENT (no file edits)
mesh-llm pi [--model <id>] [--host <h>] # writes ~/.pi/agent/models.json
goose/claudereuse a local mesh on the chosen--port.opencode/pitarget--host(default127.0.0.1:9337) and auto-start a local client only for loopback targets; the auto-started node is cleaned up when the harness exits.mesh-llm pi --write/mesh-llm opencode --writeupdate config without launching (use--hostfor remote meshes).- Agent launch commands also install available plugin skills for that agent
(
mesh-llm skills installdoes it standalone).
Manual config (any OpenAI client)
Base URL http://<host>:9337/v1, any non-empty API key:
export GOOSE_PROVIDER=openai GOOSE_MODEL="<id-from-v1-models>"
export OPENAI_HOST="http://127.0.0.1:9337" OPENAI_API_KEY="mesh"
curl -s http://localhost:9337/v1/chat/completions \
-H 'Content-Type: application/json' \
-d '{"model":"auto","messages":[{"role":"user","content":"hello"}]}'
Exact manual provider JSON for OpenCode and Pi is in docs/AGENTS.md.
Validating agent behavior
Direct API contract probe (tool-call forcing, streaming reconstruction):
scripts/qa-agent-tool-call-reliability.py \
--base-url http://127.0.0.1:9337/v1 --models auto,mesh --attempts 3 \
--output target/agent-tool-call-reliability/results.jsonl
Broader harness (models, chat, streaming, plus optional Goose/OpenCode/Pi
smokes): scripts/qa-nightly-stability.py — see docs/AGENTS.md. Use
--print-plan on either script for a side-effect-free preview.
Blackboard (cross-mesh agent coordination)
Agents can share status/questions across the mesh via the blackboard plugin — even from a client-only node:
mesh-llm plugins install blackboard
mesh-llm blackboard "STATUS: [org/repo branch:main] refactoring billing module"
mesh-llm blackboard --search "QUESTION"
MCP access: the management endpoint http://127.0.0.1:3131/mcp exposes
blackboard_post, blackboard_search, blackboard_feed. Posts are visible to
every peer — never post secrets, credentials, private paths, or customer data.
Gotchas
- Use a base URL ending in
/v1; prefer chat-completions over the Responses API unless the client documents Responses support. - Model ids must match
/v1/modelsexactly (they can contain spaces — quote them). - An empty
/v1/modelsusually means the model is still loading or no mesh was joined yet — check/api/statuson:3131(seemesh-join). - The response
"model"field tells you which node/model actually answered.