AIDE - AI Development Environment
Persistent memory, code intelligence, and multi-agent orchestration for AI coding assistants. Works with Claude Code and OpenCode.
Prerequisite: Bun — the only runtime dependency. The Go binary downloads automatically.
Install
Claude Code:
claude plugin marketplace add jmylchreest/aide
claude plugin install aide@aide
OpenCode:
bunx @jmylchreest/aide-plugin install
The Go binary downloads automatically on first run. Skills become available immediately.
What You Get
| Capability | What it does |
|---|
| Memory | Remembers preferences and context across sessions |
| Decisions | Records architectural choices, enforces them in every session |
| Code Index | Fast symbol search, call graphs, and references via tree-sitter |
| Static Analysis | Detects complexity, coupling, secrets, and code duplication |
| Survey | Maps codebase structure: modules, entry points, tech stack, churn hotspots |
| Skills | 23 built-in workflows triggered by natural language |
| Swarm | Parallel agents with full SDLC pipelines (design, test, implement, verify, docs) |
| 34 MCP Tools | Full programmatic access to all capabilities above |
Get Started
Existing project — understand the codebase:
Survey this codebase and help me understand its structure.
AIDE indexes symbols, discovers modules, tech stack, entry points, and git churn hotspots, then presents the big picture. Full guide
New project — set up guardrails:
Help me decide on the coding standards, error handling strategy, testing approach,
and architecture patterns for this project. I want to enforce SOLID, DRY, Clean Code,
and idiomatic language best practices.
The decide skill works through each topic in turn, recording separate decisions that persist across every future session. Full guide
Skills
Skills are markdown workflows triggered by keywords. Type naturally — trigger matching is fuzzy.
| Skill | Example Prompt | What Happens |
|---|
| swarm | swarm 3 implement dashboard | Parallel agents with SDLC pipeline per story |
| plan-swarm | plan swarm for the dashboard | Decomposes work into stories for swarm execution |
| decide | help me decide on auth strategy | Structured decision interview, records choices |
| design | design the auth system | Technical spec with interfaces and acceptance criteria |
| survey | survey this codebase | Maps modules, tech stack, entry points, and churn |
| test | write tests for auth | Test suite with coverage verification |
| implement | implement the feature | TDD — make failing tests pass |
| verify | verify the implementation | Full QA: tests, lint, types, build, debug artifacts |
| docs | update the documentation | Updates docs to match implementation |
| autopilot | autopilot fix all failing tests | Persistent — won't stop until verified complete |
| build-fix | fix the build errors | Iteratively fixes build/lint/type errors |
| debug | debug why login fails | Systematic debugging with hypothesis testing |
| perf | optimize the API | Performance profiling and optimization |
| review | review this PR | Security-focused code review |
| patterns | check code health | Surface code quality issues via static analysis |
| assess-findings | assess findings | Triage: read code, accept noise, keep real issues |
| code-search | find all auth functions | Search symbols, find call sites |
| memorise | remember I prefer vitest | Stores info for future sessions |
jmylchreest