context-engineering

📁 justgetai/agent-skills 📅 7 days ago
1
总安装量
1
周安装量
#51431
全站排名
安装命令
npx skills add https://github.com/justgetai/agent-skills --skill context-engineering

Agent 安装分布

windsurf 1
openclaw 1
opencode 1
cursor 1
droid 1
codex 1

Skill 文档

Context Engineering

A filesystem-based context system with autonomous workflows.

Philosophy: 80% planning, 20% execution. Each unit of work makes the next easier.


Quick Start

Full Workflow (Recommended)

/swarm-context-engineer "add stripe payments"

Runs: understand → plan → work → review → compound. Human approves at gates.

Individual Commands

/context-new feat payments    # Create spec with auto-research
/work                         # Execute current spec
/review                       # Multi-agent code review
/compound                     # Capture learnings
/status                       # See where you are

Commands

Command Purpose Auto-Agents
swarm-context-engineer Full autonomous workflow ✅ research, review
audit-context Audit codebase docs, generate specs ✅ hierarchical swarm
context-new Create spec with research ✅ repo, learnings
work Execute spec task-by-task —
review Multi-perspective code review ✅ simplicity, spec, bugs
compound Capture learnings —
status Current progress overview —
context-load Load active context —
deepen-plan Research to improve spec ✅ best practices

Directory Structure

context/
├── foundation/    # Human-authored truth (read-only for agents)
│   ├── architecture.md
│   ├── conventions.md
│   └── gotchas.md
├── specs/         # Feature/fix/improve definitions
│   └── 2026-01-26-feat-payments.md
└── tasks/         # (Optional) Atomic task breakdowns
    └── feat001-task001-setup.md

docs/
└── solutions/     # Captured learnings from /compound
    └── 2026-01-26-stripe-webhooks.md

Workflow Overview

┌─────────────────────────────────────────────────────────────┐
│  /swarm-context-engineer "feature description"             │
├─────────────────────────────────────────────────────────────┤
│  1. UNDERSTAND — load foundation, spawn research agents    │
│     ✋ Gate: confirm understanding                          │
├─────────────────────────────────────────────────────────────┤
│  2. PLAN — create spec with acceptance criteria            │
│     ✋ Gate: approve / refine / deepen                      │
├─────────────────────────────────────────────────────────────┤
│  3. WORK — execute via TodoWrite, incremental commits      │
│     ✋ Gate: ready for review?                              │
├─────────────────────────────────────────────────────────────┤
│  4. REVIEW — spawn reviewers: simplicity, spec, bugs       │
│     ✋ Gate: fix / ship / discuss                           │
├─────────────────────────────────────────────────────────────┤
│  5. COMPOUND — capture learnings to docs/solutions/        │
│     ✋ Gate: done!                                          │
└─────────────────────────────────────────────────────────────┘

Agents

Auto-spawned during workflows:

Agent When Used Purpose
repo-researcher /context-new, /let-it-rip Find relevant codebase patterns
learnings-researcher /context-new, /let-it-rip Surface past learnings
options-researcher /deepen-plan Research approaches, present options
simplicity-reviewer /review Challenge complexity
spec-reviewer /review Check spec compliance

Rules

Rule Key Point
planning 80% planning, 20% execution
foundation Foundation is read-only for agents
learnings Always run /compound after features
subagents Spawn parallel agents for research

See rules/ for complete guidelines.


Integrations

Integration Description
Linear Bidirectional sync with Linear issues

Templates


Philosophy

Compound Engineering: Each unit of work should make subsequent units easier.

  • Plan thoroughly before writing code
  • Review to catch issues and capture learnings
  • Codify knowledge so it’s reusable
  • Keep quality high so future changes are easy

Inspired by compound-engineering, adapted for filesystem-based context management.