Skip to content

Security: jayehmke/techlead

Security

SECURITY.md

Security Policy

Supported Versions

Version Supported
1.x.x
< 1.0

Security Model

Techlead is designed as a local-first development tool with the following security characteristics:

Local Execution Only

  • Runs entirely on your local machine
  • No cloud services or external data transmission
  • Knowledge database stored locally in .techlead/

Network Exposure

  • HTTP server binds to localhost by default
  • MCP mode uses stdio (no network)
  • No authentication required (local tool assumption)

Data Storage

  • SQLite database with no encryption at rest
  • Embeddings stored alongside text content
  • Session history stored in plaintext JSONL

External Dependencies

  • Requires local Ollama instance for embeddings
  • Claude Code integration via MCP protocol
  • No external API calls for core functionality

Security Considerations

DO NOT use techlead to:

  • Process sensitive credentials or secrets
  • Store PII or confidential information in decisions/learnings
  • Expose the HTTP server to public networks

Recommended Practices

  1. Run only in trusted development environments
  2. Keep .techlead/ in .gitignore to avoid committing local data
  3. Use firewall rules if running HTTP server in shared environments
  4. Regularly backup or clear .techlead/knowledge.db if it contains sensitive context

Reporting a Vulnerability

If you discover a security vulnerability:

  1. Do not open a public issue
  2. Email security concerns to the repository maintainer
  3. Include:
    • Description of the vulnerability
    • Steps to reproduce
    • Potential impact
    • Suggested fix (if any)

We aim to respond to security reports within 48 hours and will work with you to understand and address the issue.

Scope

This security policy applies to:

  • The techlead binary and all subcommands
  • The MCP server integration
  • The HTTP server endpoints
  • The knowledge database and its contents

Third-party dependencies (Ollama, sqlite-vec, etc.) have their own security policies.

There aren’t any published security advisories