load-mr-context
Pass
Audited by Gen Agent Trust Hub on Jun 26, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill uses Model Context Protocol (MCP) tools (
mcp__plugin_gitlab_gitlab__*) for all its operations, ensuring data access is mediated through an authorized and structured interface.\n- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests untrusted data from GitLab Merge Requests.\n - Ingestion points: SKILL.md implementation steps 3a, 3b, and 3d (retrieving MR descriptions, diffs, and review notes).\n
- Boundary markers: Absent; the skill does not instruct the agent to use specific delimiters or ignore instructions found within the retrieved MR data.\n
- Capability inventory: The skill's capabilities are limited to reading GitLab metadata, viewing diffs, and checking pipeline statuses via MCP tools.\n
- Sanitization: There is no explicit sanitization or filtering of the fetched external content before it is processed by the agent.\n- [SAFE]: The instruction to gather context 'silently' is used for concise UI presentation and does not represent an attempt to conceal malicious activity.
Audit Metadata