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)
17 lines
1.1 KiB
JSON
17 lines
1.1 KiB
JSON
[
|
|
{
|
|
"prompt": "Hello Hermes. We need to discuss the dev loops. you need to autoamate them such that they don't need llm inference to continue operating.",
|
|
"chosen": "Hey Alexander. Good topic. Let me pull up context on what we've been doing with the dev loops so I'm not working from stale assumptions.",
|
|
"session": "session_20260316_174724_f9091d.json"
|
|
},
|
|
{
|
|
"prompt": "Observe the tmx session timmy-loop. There is a dashbaord there that you built, it should be source controlled. There is a deep triage and a fast traige loop. Look at up. The deep traige is the perfect scope for the llm. The llm for me is always falsework and along the edges. The pipeline should be as deterministic as possible, with room for review only when anomalies arise in the system.",
|
|
"chosen": "",
|
|
"session": "session_20260316_174724_f9091d.json"
|
|
},
|
|
{
|
|
"prompt": "Maybe point to a ticket but that's it. Should be a hard coded system prompt so it doesn't need hermes.",
|
|
"chosen": "Got it. Clear direction. Let me note this and get to work.",
|
|
"session": "session_20260316_174724_f9091d.json"
|
|
}
|
|
] |