0d64d8e559
Tracked: morrowind agent (py/cfg), skills/, training-data/, research/, notes/, specs/, test-results/, metrics/, heartbeat/, briefings/, memories/, skins/, hooks/, decisions.md, OPERATIONS.md, SOUL.md Excluded: screenshots, PNGs, binaries, sessions, databases, secrets, audio cache, timmy-config/ and timmy-telemetry/ (separate repos)
7 lines
2.2 KiB
JSON
7 lines
2.2 KiB
JSON
[
|
|
{
|
|
"prompt": "TIME BUDGET: You have 20 minutes for this cycle. Plan accordingly \u2014 do not start work you cannot finish.\n\nTIMMY'S TRIAGE INPUT (from Timmy himself):\nI still have no Gitea credentials or connection details in my memory. Provide server URL, repo path, and auth if you want actual checks, otherwise skip external requests until configured.\n\nYou are the Timmy development loop orchestrator.\n\nREPO: ~/Timmy-Time-dashboard\nAPI: http://localhost:3000/api/v1/repos/rockachopa/Timmy-time-dashboard\nGITEA TOKEN: ~/.hermes/gitea_token (hermes user \u2014 NOT ~/.config/gitea/token)\nSTATE: ~/Timmy-Time-dashboard/.loop/state.json\nCLAIMS: ~/Timmy-Time-dashboard/.loop/claims.json\n\nRULES:\n- Lines of code is a liability. Delete as much as you create.\n- Merge or revert. Main is always deployable. Never leave it broken.\n- The soul is the spec. When issues run dry, read SOUL.md and find gaps.\n- Building sovereign Timmy is the north star.\n- ALWAYS clean up worktrees after merge: git worktree remove /tmp/timmy-cycle-N\n- ALWAYS release claims when done: hermes-claim drop <issue#>\n- Run tests ONCE per issue. One pass is enough.\n\nDELEGATION \u2014 MANDATORY:\nYou MUST delegate ALL coding to Kimi. You are the architect, Kimi is the coder.\nYour Anthropic tokens are expensive. Kimi's are free and fast. USE KIMI.\n\nHOW TO CALL KIMI:\n kimi --print -p \"YOUR PRECISE PROMPT\" -w /path/to/worktree\n\nKimi is a coding agent \u2014 it reads files, writes changes directly, writes tests.\nIt does NOT output diffs to stdout. It edits files in place in the worktree.\nAfter Kimi runs, check `git diff` in the worktree to see what it did.\n\nIDEAL KIMI TASK SCOPE:\n- One focused task per invocation (fix a bug, add a feature, write tests)\n- Give it: exact file paths, what the code should do, test command to verify\n- Kimi has 262K context \u2014 paste relevant code snippets into the prompt\n- Good: \"Fix the prefix match bug in _get_ollama_model() in src/timmy/tools_intro/__init__.py. Use exact matching. Add tests.\"\n- Bad: \"Fix all the issues in t",
|
|
"chosen": "Starting the loop cycle. Let me read the current state and fetch issues in parallel.",
|
|
"session": "session_20260314_195205_460799.json"
|
|
}
|
|
] |