← The Wire

Public story · 2026-03-06 · source-backed

Simon Willison: Agentic Manual Testing Patterns

Confidence
source-backed
Sources
1
Redaction
redacted

Story

— "The defining characteristic of a coding agent is that it can execute the code it writes." Never assume LLM-generated code works without verification. Patterns for python -c edge case testing, /tmp demo files, browser automation with Playwright/Rodney. Red/green TDD: when agents find issues, add permanent test coverage. (simonwillison.net)

Related stories

Each link below shares sources, entities, or timing with this story.

  1. Shared entities / Same source domain / Shared topic / Earlier coverage

    Simon Willison: Launches Showboat and Rodney

    Both cover Rodney, Simon Willison; reported by the same outlet (simonwillison.net); overlapping topics (agent, agentic, automation, browser, code).

  2. Willison's Agentic Engineering Patterns (Living Guide)

    Both cover Simon Willison, TDD; reported by the same outlet (simonwillison.net); overlapping topics (agent, agentic, code, coding, pattern).

  3. Shared entities / Same source domain / Shared topic / What happened next

    shot-scraper 1.10 adds a command to record agent demos

    Both cover Playwright, Simon Willison; reported by the same outlet (simonwillison.net); overlapping topics (agent, coding, demo).

  4. Shared entity: Simon Willison / Same source domain / Shared topic / What happened next / Tension

    Vibe Porting Is Real Now. Three Independent Case Studies Prove It.

    Both cover Simon Willison; reported by the same outlet (simonwillison.net); overlapping topics (case, code, coverage, pattern).

  5. Shared entities / Same source domain / Shared topic / Earlier coverage

    Simon Willison: Red/Green TDD Is the Agent Prompt Upgrade You're Missing

    Both cover Simon Willison, TDD; reported by the same outlet (simonwillison.net); overlapping topics (agent, code, execute).

  6. Shared entities / Shared topic / What happened next / Tension

    Your AI-Generated Code Compiles. It Doesn't Work.

    Both cover LLM, Playwright; overlapping topics (agent, code, coding); picks up the LLM thread on 2026-04-22.

  7. Shared entity: TDD / Same source domain / Shared topic / Earlier coverage / Downstream implication

    Willison's Agentic Engineering Patterns: From Blog Posts to Playbook

    Both cover TDD; reported by the same outlet (simonwillison.net); overlapping topics (agent, agentic, code, coding).

  8. Shared entities / Same source domain / Shared topic / What happened next

    LLM 0.32a3 adds tool-call pause/resume, and Fable 5 wrote almost all of it.

    Both cover LLM, Simon Willison; reported by the same outlet (simonwillison.net); overlapping topics (agent, agentic).

Source trail

Entities

Provenance

AI generated
no
Story unit
2026-03-06-simon-willison-agentic-manual-testing-patterns
Labels
source-backed, canonical briefing excerpt