Closed
Conversation
- Rewrite llmCall() to use stream:true with SSE parsing and idle timeout - Support direct llama-server (AEGIS_LLM_URL) and cloud provider (AEGIS_LLM_BASE_URL + API key) endpoints, bypassing gateway - Handle reasoning_content from thinking models (Qwen3.5) - Add progress logging every 100 tokens - Update banner and healthcheck to show actual LLM endpoint
feat: streaming SSE for benchmark LLM calls with direct endpoint support
- Replace raw fetch + manual SSE parsing with OpenAI SDK streaming - Create llmClient/vlmClient instances with proper baseURL resolution - SDK handles max_completion_tokens, auth, and SSE parsing automatically - Add openai npm dependency to skill package.json - Healthcheck uses SDK non-streaming call
refactor: use OpenAI SDK for streaming LLM calls
- Change install field from 'none' to 'npm' - Document AEGIS_LLM_URL, AEGIS_LLM_API_TYPE, AEGIS_LLM_MODEL, AEGIS_LLM_API_KEY, AEGIS_LLM_BASE_URL, AEGIS_VLM_MODEL env vars - Add openai SDK to requirements
docs: update SKILL.md with npm dependency and new env vars
A single token is often insufficient for the model to produce a valid response token, causing false ping failures.
fix: increase benchmark ping max_completion_tokens from 1 to 5
- Register yolo-detection-2026 in skills.json with FPS presets and model sizes - Rewrite SKILL.md: model_size selection (nano/small/medium/large), FPS presets - Rewrite detect.py: model_size→ultralytics mapping, frame_id echo, AEGIS_SKILL_PARAMS - Create deploy.sh: zero-assumption bootstrapper (Python/conda/pyenv, CUDA/ROCm/MPS/CPU) - Per-backend requirements files (cuda/rocm/mps/cpu) - Create config.yaml with default parameters - Create docs/detection-protocol.md: full JSONL protocol spec, bbox format, backpressure
- Add auto_start boolean to YOLO 2026 config.yaml - Expand skill-development.md: config.yaml schema format, deploy.sh, environment variables, skills.json registration, status values
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
No description provided.