Hexacore Agent
Makes coding agents prove work with your own tests, checkpoints and learned lessons
- Pricing
- Open source
- Platforms
- CLI
- Runs
- Locally
- Version
- 3.3.0
- Listing updated
- 3 days ago
About Hexacore Agent
Hexacore Agent is an open-source MCP server for developer tools. It runs locally on your machine and is started by your AI client.
Features are scaffolded as vertical slices that follow your project's layout, for 11 stacks, with a dryRun preview before any file is written. A checkpoint is taken before every write, and check_architecture fails if a slice imports from an outer layer.
Quality runs use your stack's own test runner. Each red-to-green fix becomes a lesson that earns credit only when the actual fix matches it, and lessons are kept in a library at ~/.hexacore. Tools include resume_session, execute_scaffolding, run_quality_assurance, record_lesson and record_decision.
It needs Node 20 or newer and reads the workspace from the client's MCP roots.
Install
Run in a terminal
claude mcp add hexacore-agent -- npx -y hexacore-agentAdd to claude_desktop_config.json
{
"mcpServers": {
"hexacore-agent": {
"command": "npx",
"args": [
"-y",
"hexacore-agent"
]
}
}
}Add to ~/.cursor/mcp.json
{
"mcpServers": {
"hexacore-agent": {
"command": "npx",
"args": [
"-y",
"hexacore-agent"
]
}
}
}Add to .vscode/mcp.json
{
"servers": {
"hexacore-agent": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"hexacore-agent"
]
}
}
}Add to ~/.codeium/windsurf/mcp_config.json
{
"mcpServers": {
"hexacore-agent": {
"command": "npx",
"args": [
"-y",
"hexacore-agent"
]
}
}
}Add to your client's MCP settings (most use this shape)
{
"mcpServers": {
"hexacore-agent": {
"command": "npx",
"args": [
"-y",
"hexacore-agent"
]
}
}
}Configuration
HEXACORE_WORKSPACEProject directory (default: the client's MCP roots, else the working directory)HEXACORE_HOMELesson library location (default: ~/.hexacore)
Tools this server provides (12)
Taken from the project documentation; the running server may differ.
- resume_session
- Resumes a session.
- execute_scaffolding
- Scaffolds a feature.
- run_quality_assurance
- Runs the stack's tests.
- check_architecture
- Audits architecture.
- record_lesson
- Records a lesson.
- record_decision
- Records a decision.
- design_feature
- Plans a feature.
- define_contract
- Defines a contract.
- orchestrate_task
- Orchestrates a task.
- remember_preference
- Stores a preference.
- forget_preference
- Removes a preference.
- record_handoff
- Records a handoff.
What it does
- Vertical slices
- One folder per feature.
- Checkpoints
- Saved before every write.
- Real test runs
- Uses your own runner.
- Lessons
- Learned from fixes.
- Architecture check
- Fails on outer-layer imports.
Good for
- Scaffold a feature with a dry run
- Recover after a bad edit
- Run tests inside an agent loop
- Resume where you left off
Pricing
Open source
Free npm package
Prices change. Check rbaezc.github.io before you buy.
Hexacore Agent FAQ
How many stacks?
11.
What does it need?
Node 20 or newer.
Where are lessons stored?
~/.hexacore.
Which clients?
Claude Code, Cursor, VS Code, Claude Desktop.
Is Hexacore Agent free?
Yes. Hexacore Agent is open source, so you can use and self-host it for free.
How do I add Hexacore Agent to Claude Code?
Run this command in your terminal: claude mcp add hexacore-agent -- npx -y hexacore-agent
What are the best alternatives to Hexacore Agent?
The closest alternatives to Hexacore Agent are REA, Codebase Memory, GitHub, Serena MCP: the IDE for your agent and XERJ.
Alternatives to Hexacore Agent

REA
MCP serverCLI and MCP server that lets coding agents reverse engineer apps and binaries
Runs locally- Open source
- 464K GitHub stars
- Developer tools

Codebase Memory
MCP serverLocal MCP server that builds a code knowledge graph for AI coding agents
Runs locally- Open source
- 48.7K GitHub stars
- Knowledge and memory

GitHub
MCP serverOfficial MCP server connecting AI assistants to GitHub repos, issues and workflows
Not checked yet- Remote
- Open source
- 34.5K GitHub stars
- Developer tools

Serena MCP: the IDE for your agent
MCP serverMCP toolkit giving coding agents symbol-level code retrieval and editing
Runs locally- Open source
- 31.5K GitHub stars
- Developer tools

XERJ
MCP serverLocal search engine for AI agents with cited answers over your code and docs
Runs locally- Open source
- 19.2K GitHub stars
- Knowledge and memory

Ouroboros
MCP serverSpec-first layer that pins requirements and verifies results of AI coding agent work
Runs locally- Open source
- 6.4K GitHub stars
- Developer tools