load-work
Pass
Audited by Gen Agent Trust Hub on Jul 17, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: No security issues were detected. The skill performs legitimate session management by scanning for and reading specific local context files to restore agent state. It is functionally consistent with its stated purpose.
- [PROMPT_INJECTION]: The skill processes content from CONTEXT-*-llm.md files, creating a standard surface for indirect prompt injection. However, given this is the skill's primary function for session restoration and no malicious patterns were found, the risk is negligible.
- Ingestion points: Reads files in SKILL.md Phase 1 and Phase 3 via the Read and Glob tools.
- Boundary markers: None identified; the skill relies on model interpretation of the structured context file sections.
- Capability inventory: The skill is configured with access to Bash, Read, Glob, and AskUserQuestion tools, though its instructions prioritize specific file operations.
- Sanitization: Not applicable for this session state-loading use case.
Audit Metadata