COMMITS
March 8, 2026
D
fix: define schema variable in extraction method
DhruvBhatia0 committed
D
fix: add missing sys import in reviewer.py
DhruvBhatia0 committed
D
feat: switch default provider to Gemini 3.1 Pro
DhruvBhatia0 committed
D
feat: multi-provider LLM support (OpenAI, Gemini)
DhruvBhatia0 committed
B
F1=58.5%: Return-to-open-threads, synthesize grep results, test→prod tracing
bhaktatejas922 committed
B
F1=56.3%: Fix read crash, budget investigation, missed patterns
bhaktatejas922 committed
B
fix: prevent WarpGrep 400 errors from context overflow
bhaktatejas922 committed
B
Principle-based system prompt, recall boost, post-processing dedup
bhaktatejas922 committed
March 7, 2026
D
fix: ensure every review has an agent_run_id for log correlation
DhruvBhatia0 committed
D
feat: detailed Axiom telemetry events throughout review pipeline
DhruvBhatia0 committed
D
fix: use existing Axiom dataset, add structured JSON logging
DhruvBhatia0 committed
D
fix: correct axiom-py version constraint (latest is 0.10.0)
DhruvBhatia0 committed
D
feat: add Axiom telemetry for PR review pipeline
DhruvBhatia0 committed
D
Merge branch 'docs/add-sandbox-examples'
DhruvBhatia0 committed
B
Revert to freeform+extraction with baseline prompts + reliability fixes
bhaktatejas922 committed
March 6, 2026
D
docs: add sandbox examples for WarpGrep remoteCommands
DhruvBhatia0 committed
B
report_issue tool architecture: eliminate lossy extraction step
bhaktatejas922 committed
B
Strengthen review prompt: investigation-first, confirm-before-reporting
bhaktatejas922 committed
March 5, 2026
B
Sync with agentic reviewer refactor
bhaktatejas922 committed
B
Remove fragile heuristic abstractions from review pipeline
bhaktatejas922 committed
B
Add a2a-review: personality injection + HTTP review service
bhaktatejas922 committed
March 3, 2026
B
Sync github_app with updated review pipeline
bhaktatejas922 committed
B
Add top-level README with LLM routing guide
bhaktatejas922 committed
B
Fix Python WarpGrep agent parser to match Qwen3 tool call format
bhaktatejas922 committed
B
Use 90-degree angle ASCII for subagent diagram
bhaktatejas922 committed
B
Add README with subagent explainer and ASCII architecture
bhaktatejas922 committed
B
Add WarpGrep sandbox examples (Vercel + Cloudflare)
bhaktatejas922 committed
B
Single-pass agentic review: F1=63.9% (+10pp over baseline)
bhaktatejas922 committed
D
docs: add WarpGrep example suite (10 examples)
DhruvBhatia0 committed
March 1, 2026
B
Add plan mode + structured outputs to review pipeline
bhaktatejas922 committed