Convert gitgram/ and aitimemachine/ from embedded git repos (which showed as opaque submodule gitlinks on GitHub) into fully-tracked first-party content. AITM now merges its 7 vendored repos as first-party engines: - memory: beads (Go task DAG) + context7 (docs SDK), unified into saved_contexts - git: vendored C git (bundled binary) + gitnexus (in-process code intelligence) - agentic: a2a (@a2a-js/sdk) + pal (AITM-managed Python subprocess) Includes the full session's product build: Gitgram web/mobile/backend features, AITM website/CLI/MCP, the unified memory layer, engine build tooling (scripts/build-engines.sh, Dockerfile, CI), docs (README, ENGINES.md, ARCHITECTURE.md), and extensive test suites. Build artifacts, node_modules, external/, and gemini-cli are gitignored. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
20 lines
1.1 KiB
Plaintext
20 lines
1.1 KiB
Plaintext
---
|
|
title: Data Safety
|
|
---
|
|
|
|
## Prompt Injection and Malware Pattern Detection
|
|
|
|
Context7 indexes documentation from public and private sources. To prevent malicious content from reaching AI assistants, Context7 employs a **layered malicious content detection system**.
|
|
|
|
<Frame>
|
|

|
|
</Frame>
|
|
|
|
- **Robust Detection** — Content is analyzed using a classifier tailored for Context7 to identify prompt injection attempts and malware-related patterns
|
|
- **Targeted Validation** — Suspicious content is subjected to additional checks
|
|
- **Skills.md Handling** — A separate but related detection pipeline is used for [Skills](/skills) files, adapted to their distinct structure and intent
|
|
- **Continuous Monitoring** — Flagged content is tracked and reviewed on an ongoing basis
|
|
- **Regular Updates** — Detection logic is updated to address evolving attack methods and new injection techniques
|
|
|
|
This ensures that documentation retrieved through Context7 is safe to consume by both human developers and AI coding agents.
|