pm-skills
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill collection exhibits a vulnerability surface for indirect prompt injection due to its data processing workflows.
- Ingestion points: The
meeting-analyzercomponent ingests meeting transcripts in various formats (.txt,.vtt,.srt,.docx,.json), and thescrum-mastercomponent processes sprint data from JSON exports. These files may contain instructions intended to influence the agent's behavior. - Boundary markers: There are no specific delimiters or warnings in the
SKILL.mdinstructions that direct the agent to treat input from these files as untrusted or to ignore embedded instructions. - Capability inventory: The skill is designed for use with the Atlassian MCP server, providing access to powerful tools for creating and updating issues and pages (
mcp__atlassian__create_issue,mcp__atlassian__create_page, etc.). - Sanitization: No logic or prompt-level instructions are present to sanitize or escape data extracted from external transcripts or JSON files before it is processed or used to generate API calls.
Audit Metadata