pullrequests Search Results · language:Dune language:Python language:Python language:TypeScript language:JavaScript language:C#
Filter by
243M results
🤖 AI Security Remediation
🚨 Human approval required — please review before merging.
📊 Analysis
- Severity: MEDIUM
- Confidence: 80%
- Auto-fix allowed: False
📝 Summary
The application uses the ...
ai-remediation
needs-human-review
severity-medium
gangiiopsdev
Purpose
The lm-eval accuracy gate in .buildkite/lm-eval-harness/test_lm_eval_correctness.py decides pass/fail with a fixed
relative tolerance:
rtol = eval_config.get( rtol , DEFAULT_RTOL) # DEFAULT_RTOL ...
yongzhe2160cs
Summary
Addresses production review feedback from the V2 FreshProof smoke path:
- Add the 2026 Gasoyan et al. clinical-practice discontinuation paper to the GLP-1 source seed/freshness supplement
...
jakebutler
Nazeeee0214
M31ancholyImplementation of issue #84 with:
- ARCHETYPE_REGISTRY (6 archetypes, 2 per class)
- archetypeId on PartyMember/UnitInstance
- L3 = archetype milestone (replaces generic L3 choices)
- Reactions ...
hearn1
Summary
- learning_sources.py — _historical shared block (6 feeds: Deming Institute, Drucker Institute, Federal Register FDA,
IEEE Spectrum, Nieman Lab, Health Affairs); 2 domain-specific historical ...
steventran-medtech
💡 What: Changed STOP_WORD_REGEXES from an array of 39 individual RegExp objects into a single, compiled RegExp object
(STOP_WORDS_PATTERN) constructed with NAME_STOP_WORDS.join( | ). The string replacement ...
fuegoservicios-lab
Summary
Implemented the Module 8 Integration RAG service. The pipeline retrieves top-k context from Weaviate using hybrid search
with alpha=0.5, builds the required context-injected prompt, generates ...
afrah24ali
Summary
Auto-generated ECC bundle from repository analysis.
What This Does
Merging this PR adds repo-local ECC artifacts for both Claude Code and Codex. The generated bundle captures repository
patterns, ...
ecc-tools[bot]