skills/yashas-30/nyx/internal-comms/Gen Agent Trust Hub

internal-comms

Pass

Audited by Gen Agent Trust Hub on Jun 23, 2026

Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
  • [PROMPT_INJECTION]: The skill's primary workflow involves the agent reading and summarizing content from potentially untrusted sources such as Slack threads, emails, and external press articles. These ingestion points represent an indirect prompt injection surface where adversarial instructions hidden in the processed data could influence the agent's behavior. The instructions currently lack explicit boundary markers or sanitization steps to protect the agent from malicious content in these sources.
  • Ingestion points: Slack messages, Google Drive documents, emails, calendar events, and external press articles (identified in examples/3p-updates.md, examples/company-newsletter.md, and examples/faq-answers.md).
  • Boundary markers: Absent from the prompt instructions.
  • Capability inventory: No subprocess calls, executable code, or direct network operations identified in the skill's own content.
  • Sanitization: No explicit validation or filtering instructions are provided for the gathered data.
  • [NO_CODE]: The skill is composed entirely of Markdown-based instructions, guidelines, and templates. It does not include any executable scripts (Python, Node.js, Shell), configuration files, or external binary dependencies.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 23, 2026, 10:02 AM
Security Audit — agent-trust-hub — internal-comms