Replace internal task tracking (TODO.md, FINDINGS.md) with structured documentation in docs/. Trim CLAUDE.md to agent instructions only. Co-Authored-By: Virgil <virgil@lethean.io>
1.5 KiB
1.5 KiB
CLAUDE.md
Module: forge.lthn.ai/core/go-help
Commands
go test ./... # Run all tests
go test -v -run Name # Run single test
go test -bench=. -benchmem ./... # Run benchmarks
go vet ./... # Lint
Architecture
See docs/architecture.md for full details.
Topic— help entry with title, body, sections, tags, related topicsCatalog— in-memory store;Add,List,Get,SearchParser— YAML frontmatter + Markdown section extractionSearch— inverted index with stemming, fuzzy, phrase, and prefix matchingStemmer— lightweight Porter-style suffix stripping (English)RenderMarkdown— goldmark with GFM + TypographerServer— six-route HTTP server (3 HTML, 3 JSON)Templates— embedded dark-theme HTML templatesGenerate— static site output with client-side JS searchIngest— CLI help text → Topic conversion
Coding Standards
- UK English
- EUPL-1.2 licence header on every new file:
// SPDX-Licence-Identifier: EUPL-1.2 go test ./...andgo vet ./...must pass before commit- Conventional commits:
type(scope): description - Co-Author:
Co-Authored-By: Virgil <virgil@lethean.io> - Test suffix pattern:
_Good(happy path),_Bad(errors),_Ugly(edge cases) - Benchmarks use
b.Loop()(Go 1.25) andb.ReportAllocs()
Reference
docs/architecture.md— full architectural referencedocs/development.md— development guide and standardsdocs/history.md— completed phases and known limitations