lintian

Evidence (6)

• 4The New Stack·9/30/2026research
  AI code review should support, not replace, human understanding

  The article argues that AI-assisted code review is useful for checking diffs, but it cannot replace the human role of building shared understanding of a system. It says teams increasingly skip meaningful review altogether, and that the real problem is reviewing the wrong artifact at the wrong time rather than model quality alone.

• 4Simon Willison·9/11/2026research
  Anthropic: Claude Code Needs Stronger Guardrails in Production

  Boris Cherny says code written by Claude should face a higher production-quality bar than human-written code. He describes Anthropic’s safeguards, including lint rules, extensive tests, Claude-driven end-to-end tests, daily fuzzing, automated code reviews, security reviews, and code refactoring, to avoid maintainability problems later.

• 4The New Stack·8/25/2026framework_update
  Verification First for AI Coding Agents

  The article argues that the real question for AI coding agents is not whether they run in a CLI or an IDE, but how teams verify the changes they produce. It recommends treating agent output as a proposed change and using layered checks such as linting, static analysis, type checks, and other reviews to ensure correctness, security, and maintainability.

• 7The New Stack·8/20/2026open_source
  Debian debates restricting LLM-assisted contributions

  Debian’s board has introduced competing proposals on whether to allow large language model (LLM)-assisted contributions, including a possible ban on direct code, documentation, and translation submissions created with generative AI. The debate centers on Debian’s stability, contributor accountability, and concerns about deskilling, provenance, and project norms.

• 5The New Stack·7/30/2026research
  Linting is not enough for agentic development

  The article argues that AI agents increase development speed, but teams need broader verification than linting to ensure correctness, security, and reliability. It recommends combining local checks with control-flow and data-flow analysis, dependency and secret scanning, and workflow review to keep agent-generated code safe and maintainable.

• 5Martin Fowler·4/2/2026research
  Harness engineering for coding agent users

  The article proposes a harness engineering mental model for coding agents to improve trust in AI-generated code. It argues for combining feedforward guides and feedback sensors, using both deterministic tooling and semantic evaluation to reduce errors before human review.