fe28f89574
New: interactive configurator at cc.bruniaux.com/context/ that generates a personalized CLAUDE.md starter kit based on team size, stack, and current setup. Multi-step flow (profile, current state, stack, results) with maturity scoring (Level 1-5), copy-to-clipboard artifacts, localStorage persistence. Guide content: - guide/core/context-engineering.md (1,188 lines, 8 sections): context budget, 150-instruction ceiling, modular architecture, team assembly, ACE pipeline, quality measurement, context reduction techniques - examples/context-engineering/ (10 templates): assembler.ts, profile-template.yaml, skeleton-template.md, canary-check.sh, ci-drift-check.yml, eval-questions.yaml, context-budget-calculator.sh, rules/knowledge-feeding.md, rules/update-loop-retro.md - tools/context-audit-prompt.md (543 lines): 8-dimension scoring /100 Navigation: guide/README.md, machine-readable/reference.yaml (24 new entries) Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2.8 KiB
2.8 KiB
Context Engineering Templates
Context engineering is the practice of deliberately designing what information Claude receives at session start — treating your CLAUDE.md and supporting files as a production system, not a one-time setup. These templates give you everything to build, measure, and maintain that system.
Files
| File | Description |
|---|---|
profile-template.yaml |
Developer profile for per-person context assembly |
skeleton-template.md |
Annotated CLAUDE.md skeleton with section-by-section guidance |
assembler.ts |
TypeScript script to build CLAUDE.md from profile + modules |
eval-questions.yaml |
20 self-evaluation questions to audit your CLAUDE.md |
canary-check.sh |
Behavioral regression test script (structural validation) |
ci-drift-check.yml |
GitHub Actions workflow for weekly context drift detection |
context-budget-calculator.sh |
Measures always-on token cost of your context configuration |
rules/knowledge-feeding.md |
Rule template for proactive context updates after sessions |
rules/update-loop-retro.md |
Session retrospective template to capture learnings |
Quick Start
New project — get a working CLAUDE.md in 3 steps:
# 1. Copy the skeleton and fill in your project details
cp examples/context-engineering/skeleton-template.md CLAUDE.md
# 2. Check your context budget (keep it under 10K tokens)
bash examples/context-engineering/context-budget-calculator.sh .
# 3. Run canary checks to validate structure
bash examples/context-engineering/canary-check.sh .
Existing project — audit and improve:
# Run the structural check
bash examples/context-engineering/canary-check.sh .
# Then use eval-questions.yaml to score your CLAUDE.md manually
# Target: 16+ / 20
Team setup — per-developer profiles:
# Install dependencies for the assembler
npm install js-yaml @types/js-yaml ts-node typescript
# Copy the profile template and customize
cp examples/context-engineering/profile-template.yaml .claude/profiles/yourname.yaml
# Edit .claude/profiles/yourname.yaml with your stack and preferences
# Assemble your CLAUDE.md
ts-node examples/context-engineering/assembler.ts \
--profile .claude/profiles/yourname.yaml \
--modules .claude/modules \
--output CLAUDE.md
Ongoing maintenance:
# Weekly: run canary checks
bash examples/context-engineering/canary-check.sh .
# After sessions: use the retro template
# See rules/update-loop-retro.md
# Add the CI workflow for automated drift detection
cp examples/context-engineering/ci-drift-check.yml .github/workflows/context-drift.yml
Guide Section
Full methodology and principles: guide/core/context-engineering.md
The templates here are the operational layer — the guide explains the reasoning behind each design decision.