Skip to content

Latest commit

 

History

History
71 lines (50 loc) · 3.76 KB

CLAUDE.md

File metadata and controls

71 lines (50 loc) · 3.76 KB

Walter

An engineering persona for AI agents. Skill + command framework that teaches principal engineering discipline.

Commands

npm run dev             # Astro dev server (via @walter/site)
npm run build           # Full chain: sync → optimize → plugin → copy → site → banner
npm run build:plugin    # Plugin build only (via @walter/plugin)
npm run build:site      # Astro site build (via @walter/site)
npm run lint            # Lint all: markdownlint + ESLint + astro check
npm run format          # Prettier
npm run clean           # Remove packages/*/dist/

Docs

  • DEVELOP.md — Architecture, build system, source format, adding content, development workflow
  • CONTRIBUTING.md — Setup, making changes, PRs, code style
  • SECURITY.md — Vulnerability reporting
  • README.md — User-facing: what Walter is, installation, usage

Status

  • 1 skill, 9 commands, 6 agents, 12 references — all functional, .walter/ persistence integrated
  • Build system: 4 providers, unified transforms, validation, placeholders, conditionals, auto-copy, ZIPs
  • Site: Astro, OG image, SEO meta, sitemap, favicons, 404
  • Plugin: marketplace.json + plugin.json ready
  • Quality: lint + build pass clean, Husky pre-commit hooks enforced
  • CI: GitHub Actions on PR and push to main
  • Repo: README, DEVELOP.md, CONTRIBUTING.md, SECURITY.md, issue/PR templates — all in place
  • Deployed: walter.cooking live on Vercel, domain pointed, SSL active

Known Issues

  • YAML parser doesn't strip quotes (latent — round-trip works)

Next

Content audit, frontmatter descriptions, Heisenberg evolution.

Remaining Before Release

  • Content audit — Walk every skill, command, agent, reference for tone, accuracy, completeness
  • Frontmatter descriptions — Audit all command and agent descriptions for clarity. These are what users and agents see first — clear and accurate, not gimmicky
  • Heisenberg evolution — Make write-capable, expand agent definition. Walter's subconscious — same thinking and discipline, can do real work (not just read-only advisory). Deep systems thinking + execution
  • Reference optimization — Reduce token count of reference files loaded at session start. Tighten prose, cut redundancy, preserve all behavior and intent

Future

Developer Experience

  • install.sh — Interactive installer (choose provider, target dir, global vs local)

Extensions

  • Hooks in SKILL.md frontmatter (Claude Code lifecycle events)
  • Gemini extension manifest (gemini-extension.json)
  • Codex path migration (.codex/prompts/ → .agents/skills/)
  • Provider-specific rules

Bigger Bets & Deferred

  • walter-mcp — Optional MCP server with @modelcontextprotocol/server-memory, queryable graph
  • Website download API — On-demand ZIP generation per provider

Processes

The loop: scope → break down → build → verify → hand off. Load Walter (/walter), define the problem (/formula), decompose it (/prep), build with discipline (/cook), verify against intent (/purity), seal context (/stash). Use /probe when you need research and /trace when something's broken. Every significant piece of work follows this loop — the ceremony scales with the stakes, but the thinking is always the same.

Making changes: Edit packages/, run npm run build, test with the relevant provider, run npm run lint. Husky catches lint and format issues at commit time. See DEVELOP.md for architecture and source format.

Keeping docs current: CLAUDE.md is the bootstrap — status, next action, where to look. Technical detail lives in DEVELOP.md. Contribution process lives in CONTRIBUTING.md. Focus on what helps the next agent take their first action.