 opencode
|
1cb3bdc46e
|
chore: untrack leftover chat_frame.png
|
2026-08-08 20:21:04 +03:00 |
|
|
opencode
|
b3d5c5630a
|
chore: ignore playwright logs and root-level chat artifacts
|
2026-08-08 20:20:51 +03:00 |
|
|
opencode
|
0dd2e21683
|
env: .env file for API keys (gitignored), loaded by scripts and resolve_provider; key removed from bashrc
|
2026-08-08 20:20:40 +03:00 |
|
|
opencode
|
c33452edf8
|
fix: tool-result message for non-vision tools (KeyError on width); polza provider verified end-to-end (vision + mission on first attempt)
|
2026-08-08 20:18:43 +03:00 |
|
|
opencode
|
54ee734ab5
|
providers: presets for polza/openai/ollama (--provider), keys from env, gpt-5.6-luna default for polza, multimodal heuristic for cloud ids
|
2026-08-08 20:09:19 +03:00 |
|
|
opencode
|
1d7dbb7b79
|
vision: real-time perception — auto frame attach every N turns (--look-every) and on collision, frame context cap (6), digest fallback for text models
|
2026-08-08 20:06:22 +03:00 |
|
|
opencode
|
d8373a50fc
|
vision: images are the primary channel for multimodal models (digest off by default); digest is the text-only fallback; --digest/--no-digest to override
|
2026-08-08 19:44:18 +03:00 |
|
|
opencode
|
ef02e5f9ea
|
vision: real camera frames to the model (image_url), auto-detected for multimodal ollama models; --vision/--no-vision flags; verified gemma4:12b sees frames and completed the beacon mission
|
2026-08-08 19:38:21 +03:00 |
|
|
opencode
|
6a509aacc6
|
chat: mission mode (/mission, --mission) — autonomous goal runs in background with retries, corrections, state-aware retry hints; input() moved to thread so missions run while the REPL waits; configurable nudge_limit
|
2026-08-08 13:54:44 +03:00 |
|
|
opencode
|
f4cb04a47f
|
robustness: vision/depth render off the event loop (deepcopy snapshot), chat auto-reconnect + friendly errors, demo catches connection failures
|
2026-08-08 12:33:30 +03:00 |
|
|
opencode
|
346d918118
|
chat: interactive mode (natural language -> tool calls), shared LLM driver refactor, run_chat.sh; verified with gemma4:e2b and gemma4:12b
|
2026-08-08 12:27:05 +03:00 |
|
|
opencode
|
9f877339f8
|
live: real-time browser viewer (AICC-client based), sensor-built top-down map shared module, run_live.sh one-command mode
|
2026-08-08 12:12:21 +03:00 |
|
|
opencode
|
1ebccb660b
|
visual: top-down map renderer, demo --frames-dir mode (step frames + GIF + summary), example artifacts committed
|
2026-08-08 12:03:35 +03:00 |
|
|
opencode
|
c224595f49
|
scripts: run_demo.sh launcher (no python on PATH needed), friendly port-in-use check in run_bridge.sh; README updated
|
2026-08-08 11:48:16 +03:00 |
|
|
opencode
|
4ced623b55
|
lint: ruff-clean testbed (blocking file write to thread, unused vars)
|
2026-08-08 04:29:51 +03:00 |
|
|
opencode
|
132eb25c62
|
docs: testbed README (run/agent/conformance), setup.sh deps, build log, project status
|
2026-08-08 04:29:08 +03:00 |
|
|
opencode
|
8e467f22af
|
testbed: agent demo loop (LLM via OpenAI-compatible endpoints + scripted fallback + auto handoff), E2E verified over WebSocket
|
2026-08-08 04:28:31 +03:00 |
|
|
opencode
|
b9b5e23455
|
testbed: physics fixes (rest-at-contact), pytest suite (33 tests green)
|
2026-08-08 03:43:53 +03:00 |
|
|
opencode
|
a0a8b2af75
|
testbed: bridge with world tools (proprioception, vision, depth, hear, world_query, move, turn, look_at, interact) + conformance tools; 9/9 scenarios green
|
2026-08-08 03:41:43 +03:00 |
|
|
opencode
|
c19d4334d2
|
testbed: room world engine + headless raycaster renderer (bridge skeleton)
|
2026-08-08 03:41:40 +03:00 |
|