wyjasnij-typa

Pass

Audited by Gen Agent Trust Hub on May 18, 2026

Risk Level: SAFEPROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
  • [DATA_EXFILTRATION]: The skill explicitly instructs the agent to access and analyze data from internal project management and communication tools, including Slack (via slack_read_thread), Jira, Confluence, and Fireflies meeting transcripts. This data exposure is the core functionality of the skill and relies on the agent's integrated tool capabilities.
  • [PROMPT_INJECTION]: The skill presents an attack surface for indirect prompt injection (Category 8) because it ingests and processes untrusted data from multiple external project sources.
  • Ingestion points: The agent retrieves potentially attacker-controlled content from Slack threads, Jira comments, Confluence notes, and Fireflies transripts.
  • Boundary markers: There are no boundary markers or instructions provided to the agent to disregard or isolate embedded instructions within the retrieved project data.
  • Capability inventory: The skill allows the agent to generate ready-to-use responses for Slack and create drafts for Jira and Confluence based on the analyzed data.
  • Sanitization: The instructions do not include any logic for sanitizing or validating input from these external sources before the agent interpolates the content into its analysis or generated responses.
Audit Metadata
Risk Level
SAFE
Analyzed
May 18, 2026, 08:24 AM
Security Audit — agent-trust-hub — wyjasnij-typa