Alexander Whitestone
283ec330c8
feat: implement AutoLoRA continuous improvement loop ( #1105 )
...
Tests / lint (pull_request) Failing after 16s
Tests / test (pull_request) Has been skipped
Adds the weekly sovereignty retrain cycle — Timmy improves himself
without cloud AI in the decision path.
Components:
- `timmy_automations/retrain/trajectory_exporter.py` — reads session
JSONL logs and segments them into discrete conversation trajectories
- `timmy_automations/retrain/quality_filter.py` — scores trajectories
(multi-step tasks, successful tool calls, decision points) and keeps
only HIGH/MEDIUM quality examples
- `timmy_automations/retrain/training_dataset.py` — appends filtered
trajectories to a chat-format JSONL training file with deduplication
- `timmy_automations/retrain/lora_trainer.py` — triggers mlx_lm.lora
fine-tune and registers the resulting adapter in Ollama; degrades
gracefully when mlx-lm is unavailable
- `timmy_automations/retrain/training_log.py` — records per-cycle
metrics (iteration count, train loss, skill accuracy) to JSONL + MD
- `timmy_automations/retrain/retrain.py` — main orchestrator; runs the
full pipeline and is registered as a `weekly_sunday` automation
Automation registered in `automations.json` as `retrain` with
`schedule: weekly_sunday` pointing to `hermes4-14b` as the base model.
49 new unit tests covering all components.
Fixes #1105
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com >
2026-03-23 14:17:39 -04:00
f2a277f7b5
[claude] Add vllm-mlx as high-performance local inference backend ( #1069 ) ( #1089 )
...
Tests / lint (push) Has been cancelled
Tests / test (push) Has been cancelled
Co-authored-by: Claude (Opus 4.6) <claude@hermes.local >
Co-committed-by: Claude (Opus 4.6) <claude@hermes.local >
2026-03-23 15:34:13 +00:00
e482337e50
[claude] Implement Kimi delegation for heavy research via Gitea labels ( #979 ) ( #1085 )
...
Tests / lint (push) Has been cancelled
Tests / test (push) Has been cancelled
Co-authored-by: Claude (Opus 4.6) <claude@hermes.local >
Co-committed-by: Claude (Opus 4.6) <claude@hermes.local >
2026-03-23 15:14:53 +00:00
a95cf806c8
[kimi] Implement token quest system for agents ( #713 ) ( #789 )
Tests / lint (push) Has been cancelled
Tests / test (push) Has been cancelled
2026-03-21 20:45:35 +00:00
65df56414a
[kimi] Add visitor_state message handler ( #670 ) ( #699 )
...
Tests / lint (push) Has been cancelled
Tests / test (push) Has been cancelled
Co-authored-by: Kimi Agent <kimi@timmy.local >
Co-committed-by: Kimi Agent <kimi@timmy.local >
2026-03-21 18:08:53 +00:00
dc9f0c04eb
[kimi] Add rate limiting middleware for Matrix API endpoints ( #683 ) ( #746 )
Tests / lint (push) Has been cancelled
Tests / test (push) Has been cancelled
2026-03-21 16:23:16 +00:00
f7404f67ec
[kimi] Add system_status message producer ( #681 ) ( #743 )
Tests / lint (push) Has been cancelled
Tests / test (push) Has been cancelled
2026-03-21 15:13:01 +00:00
5f4580f98d
[kimi] Add matrix config loader utility ( #680 ) ( #742 )
Tests / lint (push) Has been cancelled
Tests / test (push) Has been cancelled
2026-03-21 15:05:06 +00:00
ada0774ca6
[kimi] Add Pip familiar state to agent_state messages ( #676 ) ( #738 )
Tests / lint (push) Has been cancelled
Tests / test (push) Has been cancelled
2026-03-21 14:37:39 +00:00
646eaefa3e
[kimi] Add produce_thought() to stream thinking to Matrix ( #672 ) ( #734 )
Tests / lint (push) Has been cancelled
Tests / test (push) Has been cancelled
2026-03-21 14:09:19 +00:00
2fa5b23c0c
[kimi] Add bark message producer for Matrix bark messages ( #671 ) ( #732 )
Tests / lint (push) Has been cancelled
Tests / test (push) Has been cancelled
2026-03-21 14:01:42 +00:00
62bde03f9e
[kimi] feat: add agent_state message producer ( #669 ) ( #698 )
Tests / lint (push) Has been cancelled
Tests / test (push) Has been cancelled
2026-03-21 13:46:10 +00:00
3474eeb4eb
[kimi] refactor: extract presence state serializer from workshop heartbeat ( #668 ) ( #697 )
Tests / lint (push) Has been cancelled
Tests / test (push) Has been cancelled
2026-03-21 13:41:42 +00:00
e92e151dc3
[kimi] refactor: extract WebSocket message types into shared protocol module ( #667 ) ( #696 )
Tests / lint (push) Has been cancelled
Tests / test (push) Has been cancelled
2026-03-21 13:37:28 +00:00
4ba8d25749
feat: Lightning Network integration for tool usage ( #610 )
...
Tests / lint (push) Successful in 3s
Tests / test (push) Successful in 1m16s
Co-authored-by: Kimi Agent <kimi@timmy.local >
Co-committed-by: Kimi Agent <kimi@timmy.local >
2026-03-20 13:07:02 -04:00
12582ab593
fix: stabilize flaky test_uses_model_when_available ( #456 )
...
Tests / lint (push) Successful in 3s
Tests / test (push) Successful in 1m25s
Co-authored-by: Kimi Agent <kimi@timmy.local >
Co-committed-by: Kimi Agent <kimi@timmy.local >
2026-03-19 14:39:33 -04:00
de089cec7f
[loop-cycle-524] fix: remove numpy test dependency in test_memory_embeddings ( #451 )
Tests / lint (push) Successful in 2s
Tests / test (push) Successful in 1m38s
2026-03-19 14:22:13 -04:00
2161c32ae8
fix: add unit tests for agentic_loop.py ( #421 ) ( #447 )
...
Tests / lint (push) Successful in 2s
Tests / test (push) Failing after 1m12s
Co-authored-by: Kimi Agent <kimi@timmy.local >
Co-committed-by: Kimi Agent <kimi@timmy.local >
2026-03-19 14:13:50 -04:00
1d79a36bd8
fix: add unit tests for memory/embeddings.py ( #437 )
...
Tests / lint (push) Successful in 3s
Tests / test (push) Failing after 1m5s
Co-authored-by: Kimi Agent <kimi@timmy.local >
Co-committed-by: Kimi Agent <kimi@timmy.local >
2026-03-19 11:12:46 -04:00
db7220db5a
test: add unit tests for memory/unified.py ( #353 )
...
Tests / lint (push) Successful in 4s
Tests / test (push) Has been cancelled
Co-authored-by: Kimi Agent <kimi@timmy.local >
Co-committed-by: Kimi Agent <kimi@timmy.local >
2026-03-18 21:23:03 -04:00
