fabric-rlm
Pass
Audited by Gen Agent Trust Hub on Jul 25, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill acts as a coordinator that interpolates untrusted data from the user and from child agent results into prompts for further agent execution, creating a vulnerability surface for indirect prompt injection.
- Ingestion points: The ́π.task variable and the results from child agents in SKILL.md are used to build prompts for sub-agents and the final synthesis agent.
- Boundary markers: The skill uses basic structural headers like 'Task:' and 'Objective:' but lacks strong delimiters or explicit instructions for child agents to ignore instructions embedded within the processed content.
- Capability inventory: The agents spawned by this skill are granted access to 'read', 'grep', 'find', and 'ls' tools, allowing for broad file system read access.
- Sanitization: No escaping or validation is performed on the user-provided task strings before they are interpolated into the prompt templates.
Audit Metadata