atlassian-mcp
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [Prompt Injection] (SAFE): No attempts to override agent behavior or bypass safety filters were detected. The instructions use standard natural language for role definition.- [Data Exposure & Exfiltration] (SAFE): The skill explicitly forbids hardcoding API tokens or secrets in its constraints. No sensitive file access or unauthorized network operations are present.- [Unverifiable Dependencies & Remote Code Execution] (SAFE): No external package installations or remote script executions are requested. The skill mentions open-source MCP servers for reference but does not execute them.- [Indirect Prompt Injection] (LOW):
- Ingestion points: Processes Jira issues and Confluence pages via JQL/CQL queries (SKILL.md).
- Boundary markers: Missing specific delimiters for external data, but includes instructions to validate queries.
- Capability inventory: Capable of reading/writing to Atlassian APIs if the agent is provided with tools.
- Sanitization: Constraints explicitly require query validation and input sanitization of user-provided strings.- [No Code] (SAFE): This skill consists entirely of instructional markdown without executable scripts, commands, or binary files.
Audit Metadata