11 lines
1.8 KiB
Markdown
11 lines
1.8 KiB
Markdown
Gitea access: ezra user (id=19) on http://143.198.27.163:3000, member of Timmy_Foundation org. Token stored in /root/wizards/ezra/home/.env as GITEA_TOKEN. Profile: "Archivist, scribe, interpreter. Claude-Hermes wizard house. עזרא · 278"
|
|
§
|
|
Kimi API key in /root/.hermes/.env (KIMI_API_KEY). AllegroVPS (167.99.126.228) alive, key rotated. Cron: use execute_code with cron.jobs.create_job() if mcp_cronjob fails. Morning report cron: job a74d3cb199f5, daily 08:00 UTC (4am EST).
|
|
§
|
|
Local Timmy tool-call failure root cause: Hermes-4-14B on llama.cpp outputs tool calls as <tool_call> XML tags in text, but Hermes agent expects structured tool_calls in OpenAI API response. Fix: start llama-server with --jinja flag so chat template converts Hermes-format tags to OpenAI-structured tool calls. Alternative: patch run_agent.py ~line 7352 to fall back to HermesToolCallParser from environments/tool_call_parsers/hermes_parser.py when API returns no tool_calls.
|
|
§
|
|
Mac: apayne@100.124.176.28 (Tailscale). Timmy's Telegram goes through OpenClaw gateway (NOT Hermes). Hermes API server on localhost:8642 (/v1/chat/completions, model=hermes-agent). OpenClaw primary model: kimi/kimi-code. AllegroVPS: 167.99.126.228, 1CPU/2GB RAM — cannot run local inference, Ollama won't fit.
|
|
§
|
|
Gitea instance at http://143.198.27.163:3000. Ezra's user is "ezra" (id 19). Token available via $GITEA_TOKEN env var. Org/project context: "Timmy Time Nexus" -- multi-agent fleet coordinated by Alexander Whitestone.
|
|
§
|
|
Grand Timmy Uniwizard: Epic #94. All wizard identities dissolved. Timmy = one soul, cloud APIs = blind routed backends. Phase 1: #83,#84,#74,#78. Phase 2: #85,#86,#91,#88,#92. Phase 3: #95,#96,#97,#98. Phase 4: #87,#89,#90,#93. Phase 5: #99 dissolution. Allegro issues #74-79 still active (infra work). Old epic #82 closed into #94. |