model-checkpoint-manager
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill acts on user-initiated prompts regarding machine learning training, which could potentially contain malicious instructions intended to exploit the agent's capabilities.
- Ingestion points: User requests containing keywords like "model checkpoint manager" as defined in SKILL.md.
- Boundary markers: Absent; the skill does not use specific delimiters or instructions to ignore commands within the data it processes.
- Capability inventory: The skill is configured in SKILL.md with file access (Read, Write, Edit) and command execution (Bash for python and pip).
- Sanitization: Absent; there is no mention of validating or sanitizing the generated code or user inputs.
Audit Metadata