| 120 |
fix(llm): adapt Qwen tagged tool calls |
None |
open |
|
2026-08-12
|
2026-08-12
|
|
| 119 |
feat(llm): add first-class Gemini provider |
None |
open |
|
2026-08-12
|
2026-08-12
|
|
| 118 |
Warn on repeated test commands without workspace changes |
None |
open |
|
2026-08-11
|
2026-08-11
|
|
| 117 |
Suppress pdfjs Canvas/DOM warnings in text-only PDF extraction |
None |
open |
|
2026-08-11
|
2026-08-11
|
|
| 116 |
Document that container-created virtual environments are not host-portable |
None |
open |
|
2026-08-11
|
2026-08-11
|
|
| 115 |
Add an optional fail-closed release-readiness review phase |
None |
open |
|
2026-08-11
|
2026-08-11
|
|
| 114 |
Detect repeated overlapping reads of the same unchanged file |
None |
open |
|
2026-08-11
|
2026-08-11
|
|
| 113 |
Guide models from read_document to os.fs.read for source files |
None |
open |
|
2026-08-11
|
2026-08-11
|
|
| 112 |
Disable local llama-server probing while a cloud text provider is active |
None |
open |
|
2026-08-11
|
2026-08-11
|
|
| 111 |
Execute oversized pure-read batches in bounded waves |
None |
open |
|
2026-08-11
|
2026-08-11
|
|
| 110 |
Provider setup should probe streaming native-tool compatibility |
None |
open |
|
2026-08-11
|
2026-08-11
|
|
| 109 |
Preserve validated work around approval-gated batch barriers |
None |
open |
|
2026-08-11
|
2026-08-11
|
|
| 108 |
Add a first-class Gemini provider for Google's OpenAI-compatible endpoint |
None |
open |
|
2026-08-11
|
2026-08-11
|
|
| 107 |
Agent stalls at max_steps instead of reserving a final response |
None |
open |
|
2026-08-11
|
2026-08-11
|
|
| 106 |
OpenAI retry handling ignores Gemini error.details[].retryDelay metadata |
None |
open |
|
2026-08-11
|
2026-08-11
|
|
| 105 |
OpenAI-compatible Qwen tagged tool calls are treated as assistant text |
None |
open |
|
2026-08-11
|
2026-08-11
|
|
| 104 |
maxParallelToolCalls=1 still sends parallel_tool_calls: true to OpenAI-compatible providers |
None |
open |
|
2026-08-11
|
2026-08-11
|
|
| 103 |
OpenAI SSE tool calls without index are merged when each event has a distinct id |
None |
open |
|
2026-08-11
|
2026-08-11
|
|
| 89 |
feat(providers): add OrcaRouter preset |
None |
open |
|
2026-08-10
|
2026-08-10
|
|
| 38 |
Add any Hugging Face GGUF model, not just the curated catalog |
None |
open |
|
2026-07-26
|
2026-08-06
|
|
| 34 |
An Atomic Workflow for Context-Limited Local LLMs |
None |
open |
|
2026-07-25
|
2026-08-03
|
|
| 33 |
Feature request: Hindsight as memory implementation |
None |
open |
|
2026-07-25
|
2026-08-03
|
|
| 27 |
Feature request: Allow users to point to a model on HuggingFace in the initial set up |
None |
open |
|
2026-07-23
|
2026-07-25
|
|
