Refactoring belongs to the review stage, not the implementation loop.
Remove the refactor rule and the now-orphaned refactoring.md (its home is
the review skill), and point the loop's rule at the review skill so it
isn't re-added.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The red → green → refactor loop is anchored by leading words the model
already holds, so the step-by-step Workflow mostly restated the loop and
duplicated the horizontal-slicing anti-pattern. Drop the Workflow and
per-cycle checklist; fold their durable idea (vertical slices / tracer
bullets) into Anti-patterns and a short Rules-of-the-loop list.
Introduce **seam** as the leading word for where tests go, collapsing the
old Philosophy "public interfaces" prose and the Planning "confirm
interface / behaviors" handshake into one rule: test only at pre-agreed
seams, confirmed with the user before any test is written.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Tests whose assertion is recomputed the way the code computes it pass by
construction and give zero confidence. Add it as a peer of the existing
implementation-coupling anti-pattern: a Philosophy principle, a per-cycle
checklist gate, and a BAD/GOOD example pair in tests.md. Includes a patch
changeset.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
- Introduced a human-in-the-loop script for bug diagnosis to capture user feedback.
- Created ADR and CONTEXT formats to standardize architectural decision records and domain context documentation.
- Developed a domain modeling skill to refine terminology and maintain a shared language.
- Enhanced the improve-codebase-architecture skill to provide visual reports and deepening opportunities.
- Consolidated grilling and handoff skills for better user interaction and documentation.
- Updated existing skills to improve clarity and consistency in descriptions and functionality.
Engineering skills no longer hard-code GitHub or specific label strings.
A new setup skill scaffolds an `## Agent skills` block in
AGENTS.md/CLAUDE.md plus `docs/agents/` so each repo can declare its own
backlog backend, triage label vocabulary, and domain doc layout. Skills
that need the mapping (to-issues, to-prd, triage) point at the setup
skill; skills that only soften with it (diagnose, tdd,
improve-codebase-architecture, zoom-out) stay vague. ADR-0001 records
the split.
Closes#88, #89.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- Introduced TDD skills including deep modules, interface design, mocking, refactoring, and testing guidelines.
- Added skills for breaking plans into GitHub issues and creating PRDs from conversation context.
- Implemented productivity skills for scaffolding exercises, setting up pre-commit hooks, and managing notes in Obsidian.
- Created a caveman communication mode for concise technical responses and a grilling technique for thorough plan discussions.
- Developed a skill for writing new agent skills with structured templates and guidelines.
- Included git guardrails to prevent dangerous git commands and a migration guide for using @total-typescript/shoehorn in tests.
