AI Coding Live News

MarkuprPlus open-sources desktop/CLI/MCP bug capture: window recording plus on-device Whisper to agent Markdown

@DanKornas highlighted MarkuprPlus (github.com/hashfunction/MarkuprPlus), an MIT desktop app, CLI, and MCP that turns window-locked screen recordings plus on-device Whisper narration into structured Markdown (one annotated screenshot per mark) for Claude Code, Cursor, Codex, and Windsurf. Local-first session folders keep reports, screenshots, video, audio, and metadata together.

MarkuprPlus open-sources desktop/CLI/MCP bug capture: window recording plus on-device Whisper to agent Markdown

Key Takeaways

  • Each on-screen mark yields its own annotated screenshot and issue.
  • Narration is transcribed on-device with Whisper.
  • Desktop, CLI, and MCP; MIT; hand off to Claude Code, Cursor, Codex, Windsurf.
Read full story

skill-audit scans agent skills, plugins, and MCP configs for supply-chain risk

@AverageAiBro highlighted ondrej-merkun/skill-audit, a local scanner for prompt-injection and malware patterns across Claude Code, Codex, Copilot, Cursor, Gemini CLI, Windsurf, Cline, AGENTS.md, installed skills, plugins, and MCP configs. It ships 46 rules with no cloud upload, optional deeper review through localhost OpenAI-compatible models, CI support via --fail-on REVIEW, and a GitHub Action under the @ondrej-merkun/skill-audit npm package.

Key Takeaways

  • Scans skills, plugins, MCP configs, and AGENTS.md locally across major coding-agent clients.
  • Forty-six rules detect prompt injection and malware patterns without uploading configs to the cloud.
  • Supports --fail-on REVIEW, a GitHub Action, and optional localhost model review for install and CI gates.
Read full story

Auto-Guard open-sources Jev-backed allow/escalate/block gates for coding-agent tool calls

@roshanchandna introduced rchandnaWUSTL/auto-guard: before a coding agent runs a tool call, Auto-Guard sends the proposal to TypeSafe Jev and returns allow, escalate (ask you with a one-line LLM reason), or block. Install with pip install agent-autoguard then autoguard install for Claude Code by default, with adapters for Cursor, Codex, Copilot, Gemini, and Windsurf. The author cites roughly 0.4s and about $0.000025 per check.

Key Takeaways

  • Pre-tool-call gate with Jev verdicts: allow, escalate for human approval, or block.
  • pip install agent-autoguard then autoguard install for Claude Code; adapters cover Cursor, Codex, and peers.
  • Author cites ~0.4s and ~$0.000025 per check; rchandnaWUSTL/auto-guard publishes code and known limits.
Read full story

Codeium Windsurf introduces Cascade Hybrid Orchestration with adaptive routing between DeepSeek-R1 and Claude 3.7

Codeium rolled out Cascade Hybrid Orchestration for its Windsurf IDE. The engine routes architectural planning and algorithmic decomposition to DeepSeek-R1 while switching to Claude 3.7 Sonnet for rapid code writing and patch application, boosting composite task completion rates by 28%.

Key Takeaways

  • Dual-engine workflow: DeepSeek-R1 for architectural planning and Claude 3.7 for code generation.
  • Boosts complex multi-step debugging resolution rates by 28% while reducing rollback retries.
  • Available by default for Pro and Team subscribers in the latest Windsurf IDE update.
Read full story
ADSponsored

Codeium Windsurf Cascade introduces persistent workflow memory graphs to retain project architectural conventions

Codeium launched Cascade Memories for its Windsurf IDE. The agent automatically distills team conventions, API paradigms, deprecated methods, and architectural patterns into a persistent knowledge graph, applying learned rules across sessions and branches without manual prompting.

Key Takeaways

  • Automatically extracts team coding standards and design constraints from interactions.
  • Memories exportable as team config files (.windsurf/rules) shared across Git repos.
  • Prevents repeated mistakes by preserving architectural invariants across sessions.
Read full story
Linked Models & Agents:

Cognition Unveils Devin Desktop as Next-Generation Windsurf with Deep Autonomous Workflow Integration

Cognition announced Devin Desktop, integrating Windsurf's IDE foundation with Devin's multi-agent command center and Agent Client Protocol (ACP) for unified local and cloud agent orchestration.

Key Takeaways

  • Windsurf IDE evolves into Devin Desktop, providing a centralized cockpit for local and cloud agent fleets;
  • Native Agent Client Protocol (ACP) support enables Codex, Claude Agent, and OpenCode to share project context;
  • Preserves existing Windsurf keybindings, extensions, and LSP configurations; free to download at devin.ai.
Read full story
Linked Models & Agents:

Codeium Unveils Windsurf Enterprise: Self-Hosted Cloud Clusters & SOC2 / HIPAA Compliance

Codeium released Windsurf Enterprise Edition supporting air-gapped on-premise VPC clusters and SOC2 compliance.

Key Takeaways

  • Air-gapped on-premise deployment with enterprise SSO integration (Okta, Azure AD, SAML)
  • Centralized administrative console for managing team-wide Rules Hub and model routing policies
  • Native bidirectional integration with internal self-hosted GitLab, Bitbucket, and Jira
Read full story

Cognition launches Devin Desktop: Unified fleet management for local and cloud coding agents with ACP

Cognition launched Devin Desktop as the evolution of Windsurf IDE. Serving as a command center for local and cloud agent fleets, it introduces full Agent Client Protocol (ACP) support to orchestrate Codex, Claude Agent, and OpenCode on a unified surface while retaining full IDE editing for manual reviews.

Key Takeaways

  • Command surface manages and monitors dozens of concurrent local and cloud worker agents.
  • Adopts Agent Client Protocol (ACP) for multi-agent interoperability across heterogenous models.
  • Built on Windsurf IDE foundations, carrying over extensions, LSPs, and custom settings.
Read full story