iampantherr/SecureContext

7 stars · Last commit 2026-05-28

Secure memory & context optimization MCP plugin for Claude Code. Drop-in replacement for context-mode with credential isolation, SSRF protection, MemGPT-style persistent memory, and hybrid BM25+vector search. 84 security tests, zero cloud sync.

README preview

# SecureContext

> **The security and memory layer for Claude Code.** Persistent memory that survives restarts, cryptographic audit trail for every tool call, and the only HMAC-verified admission gate for Anthropic-style filesystem skills. Runs locally on PostgreSQL — zero cloud sync, MIT-licensed.

[![Version](https://img.shields.io/badge/version-0.29.0-blue)](package.json)
[![Tests](https://img.shields.io/badge/tests-786%20passed-brightgreen)](src)
[![Security Tests](https://img.shields.io/badge/security%20red%20team-60%2B%20RT%20IDs-brightgreen)](security-tests)
[![CI](https://github.com/iampantherr/SecureContext/actions/workflows/ci.yml/badge.svg)](https://github.com/iampantherr/SecureContext/actions)
[![License: MIT](https://img.shields.io/badge/License-MIT-blue.svg)](LICENSE)
[![Node](https://img.shields.io/badge/node-%3E%3D22-green)](package.json)

> ⚠️ **Note on the "SafeSkill 20/100 Blocked" PR comment:** that score is a false positive from a regex-based scanner that doesn't understand the difference between *defending against* a pattern and *using* it. See [SAFESKILL_RESPONSE.md](SAFESKILL_RESPONSE.md) for the line-by-line refutation. The actual project has 786 passing tests including 60+ red-team attack IDs verified against a real threat model.

---

## What SecureContext does for you

If you use Claude Code regularly, you've already hit these walls:

1. **Context window resets.** Every new session you re-paste the same files, re-explain the same decisions, re-discover the same gotchas. You're paying tokens for the same orientation work over and over.

View full repository on GitHub →