This release adds 3 notable features for engineering teams evaluating rollout.
✓ No known CVEs patched in this version
Summary
AI summaryAgents can onboard and bind new projects directly within the IDE using three new tools.
Full changelog
What's New
AI agents can now onboard new projects and bind to them without leaving the IDE.
New Tools (21 total)
| Tool | Purpose |
|------|---------|
| setup_project(name, git_repo_url, ...) | Register + index a new project directly from the agent |
| set_project(name) | Bind all subsequent calls to a specific project (sticky session) |
| get_active_project() | Show which project is currently bound |
How It Works
- Agent opens a new project → Flaiwheel is already connected (global MCP config)
- Agent calls
set_project("my-app")→ all subsequent calls target "my-app" - If the project isn't registered:
setup_project("my-app", git_repo_url="...")creates it and auto-binds
Resolution Order
Every tool still has the project parameter as a hard safeguard:
- Explicit
project=— always wins - Sticky
_active_project— set byset_project/setup_project - Default first project — backward compat
Agent Instructions Updated
- Cursor rules (
.cursor/rules/flaiwheel.mdc) andAGENTS.mdnow instruct agents to callset_project()at the start of every conversation - All
FLAIWHEEL_TOOLS.mdtemplates updated with the full 21-tool table
Upgrade
Run the install script from your project directory to update.
Weekly OSS security release digest.
The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.
No spam, unsubscribe anytime.
Share this release
About dl4rce/flaiwheel
Self-hosted memory and governance layer for AI coding agents. 28 MCP tools with structured knowledge capture, hybrid search (semantic + BM25 + cross-encoder reranking), behavioral documentation nudges, cold-start codebase analyzer, and git-native storage. Single Docker container, zero cloud dependencies.
Related context
Beta — feedback welcome: [email protected]