ThibautBaissac/rails_ai_agents
Command Claude Code
Creates one polished, self-contained HTML artifact from review findings, audit notes, PR feedback, code review output, security review notes, design review notes, QA reports, or implementation assessments.
ThibautBaissac/rails_ai_agents
Command Claude Code
Execute a small change by processing all tasks sequentially from tasks.md — no subagents, no hooks, no checklists.
ThibautBaissac/rails_ai_agents
Command Claude Code
Create a lightweight change specification for bug fixes and small features — skips the full SDD ceremony.
ThibautBaissac/rails_ai_agents
Command Claude Code
Generate a flat task list (3-8 tasks) for a small change based on the change spec.
ThibautBaissac/rails_ai_agents
Command Claude Code
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
ThibautBaissac/rails_ai_agents
Command Claude Code
Generate a custom checklist for the current feature based on user requirements.
ThibautBaissac/rails_ai_agents
Command Claude Code
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
ThibautBaissac/rails_ai_agents
Command Claude Code
Create or update the project constitution from interactive or provided principle inputs, ensuring all dependent templates stay in sync.
ThibautBaissac/rails_ai_agents
Command Claude Code
Execute the implementation plan by delegating each task in tasks.md to the matching specialist agent (model-agent, service-agent, rspec-agent, ...) in a fresh context, orchestrated phase-by-phase. Use /sdd:implement for inline single-context execution instead.
ThibautBaissac/rails_ai_agents
Command Claude Code
Execute the implementation planning workflow using the plan template to generate design artifacts.
ThibautBaissac/rails_ai_agents
Command Claude Code
Adversarial review of the feature spec from security, performance, edge-case, scalability, and regulatory perspectives to catch blind spots before planning.
ThibautBaissac/rails_ai_agents
Command Claude Code
Validate that the codebase implements what the feature spec promises using a 4-layer hybrid analysis — no code annotations required.
Command Claude Code
Fetch an article from a URL, create a raw source file, and ingest it into the KB.
Command Claude Code
Ingest a raw source file into the KB — creates/updates wiki pages, index, and log.
Command Claude Code
Health-check the KB for contradictions, orphans, stale content, and broken links.
Command Claude Code
Repurpose the KB for a different topic — cleans content, rewrites schema, updates tooling and viz.
Skill Claude CodeCodex
Answer a question from the KB with citations, optionally filing the answer as a new page. Use when the user asks a physics question, wants to query the knowledge base, or asks about topics covered in the KB (theories, concepts, people, experiments, open questions). WHEN NOT: Ingesting new sources (use kb:ingest)…
Instructions file
Claude Code instructions for ThibautBaissac/kb-physics, covering claude.md, what this is, commands, requires: pandoc (pdf/html/docx), python-pptx (slides) and from arguments.