async-operations
Pass
Audited by Gen Agent Trust Hub on Apr 24, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill ingests untrusted data from user code files and has the capability to modify the filesystem, creating an attack surface for indirect prompt injection attacks where malicious code could influence agent behavior.
- Ingestion points: Source code files matching the glob '**/*.{svelte,js,ts}'.
- Boundary markers: No delimiters or specific instructions to disregard embedded commands in the input data are present.
- Capability inventory: The skill is configured with Read, Write, and Edit tools.
- Sanitization: There is no evidence of input validation or sanitization before the agent processes the code.
- [COMMAND_EXECUTION]: The SKILL.md instructions include a directive for the agent to execute a shell command to retrieve context from a hidden directory.
- Evidence: The 'Memory Protocol' section instructs the agent to run 'cat .claude/context/memory/learnings.md'.
Audit Metadata