Coding Tools
GitHub Launches Duplicate Issue Detection Preview
GitHub Changeloghigh confidence
Why it matters
- →Reduces time spent on triaging duplicate issues for maintainers.
- →Enhances automation in issue management with AI tools.
- →Improves efficiency in handling large repositories.
GitHub has launched a public preview feature to help maintainers manage duplicate issues more efficiently. The new tool suggests potential duplicate issues during the creation process, displaying up to three matches inline. This aims to reduce the time spent on triaging by highlighting existing reports. Additionally, AI tools connected to the GitHub MCP server can now read and write issue fields, enabling automated triaging and filtering. This update is expected to streamline issue management for large repositories.
Read originalThe story around this
Earlier coverage that leads up to this article, and what followed. Lines connect each piece to the closest one after it, converging here.
How we got here
- 1GitHub MCP Server Offers Dependency Scanning Preview
GitHub Changelog · May 5, 2026 · Same story
- 2GitHub Issue Fields Now in Public Preview
GitHub Changelog · May 21, 2026 · Same story
Follow this story
Open the full story →GitHub Launches Duplicate Issue Detection Preview
2 developments
- Jun 18 · GitHub Changelog
GitHub Launches Duplicate Issue Detection Preview (This article) - Jul 2 · GitHub Changelog
GitHub Issue Fields Now Generally Available↳ GitHub Issue Fields are now generally available for all organizations with AI integration via MCP server support.
More in Coding Tools
Coding Toolscoding
Claude Code v2.1.316: Subagents and Gateway Budgets
Anthropic is tightening control over Claude Code's autonomous behavior with new environment variables that let you disable 'ultrathink' or cap subagent effort levels. For teams, the introduction of pooled group budgets on the gateway allows shared monthly spending limits across members, a practical fix for enterprise cost management. The release also patches critical bugs where shell commands would hang or crash on exit, stabilizing the developer experience significantly.
Claude Code Releases
Coding Toolscoding
llama.cpp b12022 adds ROCm 10 and Snapdragon support
This release quietly cements llama.cpp as the universal inference runtime by finally bringing ROCm 10.0 to Linux and Windows alongside CUDA 13.4, effectively closing the hardware gap for AMD users who were previously left behind. The inclusion of native Snapdragon support on Linux marks a significant step toward running local LLMs on ARM-based mobile chips without emulation overhead. While KleidiAI on Apple Silicon is disabled in this build, the broader expansion to diverse accelerators ensures that developers aren't locked into NVIDIA ecosystems anymore.
llama.cpp Releases
Coding Toolscoding
llama.cpp b12023 fixes shared prefix bug
This release addresses a specific inference artifact where decision prefixes were incorrectly bleeding into output embeddings. By capping shared prefixes at the first output position, llama.cpp ensures that option tokens remain distinct for parent and child tasks. This is a targeted fix for multi-step reasoning or voting workflows rather than a performance boost. It stabilizes complex generation patterns without changing the underlying model architecture.
llama.cpp Releases