dev-context-engineering

Pass

Audited by Gen Agent Trust Hub on Oct 6, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides utility scripts for repository documentation analysis (scan_context_artifacts.py, query_context_graph.py, validate_context_graph.py) which operate locally and do not perform network requests or unauthorized file modifications.
  • [SAFE]: The skill includes security-enhancing assets, such as GitHub Action workflows for PII (Personally Identifiable Information) detection and compliance rules for FCA-regulated environments.
  • [INDIRECT_PROMPT_INJECTION]: The skill's primary function is to ingest and process documentation from a target repository. This creates a surface for indirect prompt injection if the scanned files contain malicious instructions.
  • Ingestion points: scan_context_artifacts.py crawls the repository file system; validate_context_graph.py and query_context_graph.py read the generated context-graph.json.
  • Boundary markers: The scripts do not explicitly use delimiters or ignore directives for content extracted from repository files.
  • Capability inventory: The Python scripts perform file reads and JSON parsing, outputting analysis results to the console or local files.
  • Sanitization: Content is parsed using standard json.loads and regex patterns; no explicit sanitization of processed strings is present.
  • [EXTERNAL_DOWNLOADS]: The documentation references established development tools and frameworks such as GitHub Spec Kit, Anthropic Agent Skills, and OpenSpec. These are well-known industry resources and do not represent a security risk.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 6, 2026, 02:03 PM