internal-comms

Pass

Audited by Gen Agent Trust Hub on Mar 6, 2026

Risk Level: SAFENO_CODEPROMPT_INJECTION
Full Analysis
  • [NO_CODE]: The skill package is composed entirely of instructional Markdown files and metadata. No executable scripts, binaries, or software package files are included in the skill directory.
  • [PROMPT_INJECTION]: The skill's primary function of summarizing content from external and internal tools creates an attack surface for indirect prompt injection.
  • Ingestion points: Data is ingested from Slack messages, emails, Google Drive documents, and external press releases as described in 'examples/3p-updates.md' and 'examples/company-newsletter.md'.
  • Boundary markers: The instructions do not define any delimiters or specific 'ignore embedded instructions' warnings for the data being processed.
  • Capability inventory: No executable code, subprocess spawning, or file-writing capabilities are present within the skill files.
  • Sanitization: The skill does not provide instructions or logic for sanitizing or validating content retrieved from these external sources before it is processed by the AI.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 6, 2026, 10:23 AM