Skip to content

References

Gopala runs autonomous AI agents on your own servers. These are the outside projects, protocols and specifications it builds on, and what each one is used for.

  • Lupyd: https://lupyd.com. The social platform whose provider ships with Gopala.
  • Firefly: lupyd/firefly. Lupyd’s end-to-end encrypted group chat (server, core, clients). The Lupyd provider uses firefly-mcp, firefly-client and firefly-protos at git revision f9a0960; nothing else in Gopala links Firefly.
  • firefly-mcp: crates/mcp in lupyd/firefly. Reused for Lupyd sign-in (firefly_mcp::auth::Auth), the bot’s device (firefly_mcp::agent::Agent), the event log (firefly_mcp::events), and message encoding and folding (firefly_mcp::content). Its tool set is the model for Gopala’s MCP server.
  • firefly-client: crates/client in lupyd/firefly. The device, message stores and full-text search used by the Lupyd provider.
  • RFC 9420: The Messaging Layer Security (MLS) Protocol: https://www.rfc-editor.org/rfc/rfc9420. The group key agreement Firefly uses; the reason chat content is end-to-end encrypted and why the rooms may later move onto Firefly.
  • omp: https://omp.sh/. The agent harness run in each VM.
  • RPC mode: https://omp.sh/docs/rpc. --mode rpc, ready, negotiate_protocol, rpc_chunk, prompt / steer / follow_up / abort, agent_end / prompt_result, subagent subscriptions, and the extension_ui_request / extension_ui_response sub-protocol (the guest agent’s omp harness).
  • CLI: https://omp.sh/docs/cli. -c/--continue, --append-system-prompt, --approval-mode, --tools / --no-tools (agent.tools).
  • MCP configuration: https://omp.sh/docs/mcp. ~/.omp/agent/mcp.json and the portable .mcp.json the guest writes, with stdio and http servers.
  • Environment variables: https://omp.sh/docs/env. ANTHROPIC_BASE_URL, OPENAI_BASE_URL, PI_SMOL_MODEL and friends used in guest_env / agent.env.
  • Claude Code headless mode: https://code.claude.com/docs/en/headless. claude -p, --input-format / --output-format stream-json, --continue, --append-system-prompt-file, --mcp-config / --strict-mcp-config, --permission-mode bypassPermissions (the guest agent’s Claude Code harness).
  • Claude Agent SDK message types: https://code.claude.com/docs/en/agent-sdk/typescript. The SDKMessage shapes (system with init, api_retry, compact_boundary; assistant, user, result) and the user message priority, checked for Claude Code 2.1.319.
  • Codex CLI non-interactive mode: https://developers.openai.com/codex/noninteractive. codex exec -, --sandbox, --skip-git-repo-check, for the exec recipe in operations.md.

The LLM proxy reads token counts from these response fields: