pm-skills
Pass
Audited by Gen Agent Trust Hub on Nov 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill suite contains components designed to ingest and process untrusted external data, creating a surface for indirect prompt injection attacks.
- Ingestion points: The
meeting-analyzerskill processes various transcript formats including.txt,.md,.vtt,.srt,.docx, and.jsonfiles. Theteam-communicationsskill is designed to extract context from external communication platforms like Slack, Gmail, and Google Drive. - Boundary markers: The normalization workflow for transcripts in
meeting-analyzerlacks explicit mention of delimiters or instructions to the model to ignore potential directives embedded within the processed text. - Capability inventory: The skills possess significant capabilities via MCP (Model Context Protocol), including searching, creating, and updating content in Jira and Confluence, as well as executing local Python scripts for data analysis.
- Sanitization: The provided documentation does not outline specific sanitization or filtering procedures for external content before it is used to generate reports or influence agent behavior.
Audit Metadata