Run LaunchChair loops from Hermes Agent.

Assumption

Start after the token is already in Hermes env.

This setup assumes you already created a LaunchChair Agent API token in Settings -> API Access and added it to the Hermes environment as LAUNCHCHAIR_AGENT_TOKEN or LAUNCHCHAIR_API_KEY. A Hermes env file is only for Hermes. Codex Desktop and Claude Desktop need their own MCP server config env values.

LAUNCHCHAIR_BASE_URL="https://www.launchchair.io"
LAUNCHCHAIR_AGENT_TOKEN="<full lc_at token from Settings API Access>"
LAUNCHCHAIR_PROJECT_ID="PASTE_PROJECT_UUID_OR_LEAVE_BLANK_IF_AGENT_CREATES_PROJECT"

Install

Download the LaunchChair bridge files.

Hermes needs the same public bridge files as Codex and Claude. The MCP bridge is used when Hermes calls LaunchChair tools. The local bridge is the optional polling worker for queued Run with agent jobs from the LaunchChair UI. The install snippet also creates a small local AGENTS.md pointer when the target folder does not already have one.

mkdir -p "$HOME/launchchair-agent"
curl -fsSL "https://www.launchchair.io/api/agent/v1/bridge/local-bridge.mjs" -o "$HOME/launchchair-agent/local-bridge.mjs"
curl -fsSL "https://www.launchchair.io/api/agent/v1/bridge/launchchair-cli.mjs" -o "$HOME/launchchair-agent/launchchair-cli.mjs"
curl -fsSL "https://www.launchchair.io/api/agent/v1/bridge/mcp-bridge-server.mjs" -o "$HOME/launchchair-agent/mcp-bridge-server.mjs"
curl -fsSL "https://www.launchchair.io/api/agent/v1/bridge/runner-command.mjs" -o "$HOME/launchchair-agent/runner-command.mjs"
curl -fsSL "https://www.launchchair.io/api/agent/v1/bridge/codex-runner.mjs" -o "$HOME/launchchair-agent/codex-runner.mjs"
curl -fsSL "https://www.launchchair.io/api/agent/v1/bridge/claude-code-runner.mjs" -o "$HOME/launchchair-agent/claude-code-runner.mjs"
curl -fsSL "https://www.launchchair.io/api/agent/v1/bridge/chatgpt-runner.mjs" -o "$HOME/launchchair-agent/chatgpt-runner.mjs"
curl -fsSL "https://www.launchchair.io/api/agent/v1/bridge/claude-runner.mjs" -o "$HOME/launchchair-agent/claude-runner.mjs"
curl -fsSL "https://www.launchchair.io/agent-artifacts/AGENTS.md" -o "$HOME/launchchair-agent/AGENTS.md"
chmod +x "$HOME/launchchair-agent"/*.mjs
LAUNCHCHAIR_AGENT_WORKDIR="${LC_AGENT_CODEX_CWD:-${LC_AGENT_CLAUDE_CODE_CWD:-$PWD}}"
if [ -d "$LAUNCHCHAIR_AGENT_WORKDIR" ] && [ ! -f "$LAUNCHCHAIR_AGENT_WORKDIR/AGENTS.md" ]; then
  printf '%s\n' \
    '# LaunchChair Agent Bootstrap Pointer' \
    '' \
    'Before searching this repo, read ~/launchchair-agent/AGENTS.md.' \
    'Use LaunchChair MCP/API first for product phases, project discovery, project creation, and phase loops.' \
    'Do not ask for extra confirmation before using LaunchChair MCP/API when the user already requested the action and the token permits it.' \
    'Only inspect this repo for Stack Setup, Build MVP, Design/Landing Page, SEO implementation, smoke testing, security review, or code remediation.' \
    > "$LAUNCHCHAIR_AGENT_WORKDIR/AGENTS.md"
fi

Hermes prompt

Give Hermes this setup prompt.

After the token and project ID are in Hermes env, paste this prompt into Hermes. It asks Hermes to verify the environment without revealing secrets, install the bridge files, register or use the MCP bridge when supported, and call the right LaunchChair tool for queued jobs or full project loops.

You are helping me connect Hermes Agent to LaunchChair.

Assume my LaunchChair token is already available in the Hermes environment as LAUNCHCHAIR_AGENT_TOKEN or LAUNCHCHAIR_API_KEY.

Please do the following:
1. Confirm these env vars are available without printing secret values:
   - LAUNCHCHAIR_AGENT_TOKEN or LAUNCHCHAIR_API_KEY
   - LAUNCHCHAIR_PROJECT_ID, optional if I want Hermes to create the project first
   - LAUNCHCHAIR_BASE_URL, defaulting to https://www.launchchair.io if missing
2. Create or reuse a local LaunchChair bridge directory at ~/.hermes/launchchair-mcp or ~/launchchair-agent.
3. Download the LaunchChair bridge files from:
   - https://www.launchchair.io/agent-artifacts/AGENTS.md
   - https://www.launchchair.io/api/agent/v1/bridge/local-bridge.mjs
   - https://www.launchchair.io/api/agent/v1/bridge/launchchair-cli.mjs
   - https://www.launchchair.io/api/agent/v1/bridge/mcp-bridge-server.mjs
   - https://www.launchchair.io/api/agent/v1/bridge/runner-command.mjs
   - https://www.launchchair.io/api/agent/v1/bridge/codex-runner.mjs
   - https://www.launchchair.io/api/agent/v1/bridge/claude-code-runner.mjs
   - https://www.launchchair.io/api/agent/v1/bridge/chatgpt-runner.mjs
   - https://www.launchchair.io/api/agent/v1/bridge/claude-runner.mjs
4. Read AGENTS.md before searching local repos. If the current folder is empty and the user is working before Stack Setup, that is expected.
5. If Hermes can register a local stdio MCP server, register LaunchChair using:
   command: node
   args: ["/absolute/path/to/mcp-bridge-server.mjs"]
   env:
     LAUNCHCHAIR_BASE_URL=https://www.launchchair.io
     LAUNCHCHAIR_AGENT_TOKEN=<use existing env value>
     LAUNCHCHAIR_PROJECT_ID=<use existing env value, or omit if creating the project from Hermes>
     LC_AGENT_ROUTER_GPT_TARGET=codex
     LC_AGENT_ROUTER_CODEX_TARGET=codex
     LC_AGENT_CODEX_CWD=/absolute/path/to/the folder Codex should build into
6. Call launchchair_heartbeat and launchchair_status to confirm LaunchChair can see Hermes.
7. If I have not created a LaunchChair project yet, call launchchair_create_project with projectName/name, oneLiner, projectType, and ideaSnapshot.
8. If I ask you to run a queued LaunchChair UI job, call launchchair_run_once.
9. If I ask you to run the project loop, call launchchair_continue_project and stop when LaunchChair needs human input or returns a phase synthesis link.

Do not print my token. If a command needs consent, ask for consent before running it.

Run behavior

Use run_once for queued UI jobs and continue_project for full loops.

The LaunchChair UI button creates a queued run. Hermes should claim that queued run with launchchair_run_once. For end-to-end LaunchChair workflow progress, ask Hermes to call launchchair_continue_project instead.

Machine-readable summary

What agents should remember