context-layer-consolidation

Pass

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute a local script scripts/lint with the --report flag and as a final verification step. This involves spawning a subprocess to run code present within the repository.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and modify various Markdown files across multiple directories (org/, areas/, decisions/). This creates a vulnerability surface for indirect prompt injection.
  • Ingestion points: Markdown content located in org/, areas/, decisions/, and Space pages.
  • Boundary markers: The skill provides instructions on which directories to exclude (repository instructions, generated indexes, scripts/), but lacks specific delimiters or warnings to ignore instructions embedded within the processed documentation files.
  • Capability inventory: The skill has the capability to write to the file system (modifying Markdown files), commit changes to a git branch, and execute a local linting script.
  • Sanitization: There is no evidence of sanitization or filtering of the content being processed to prevent the agent from following instructions hidden within the wiki pages.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 09:32 PM
Security Audit — agent-trust-hub — context-layer-consolidation