xy-script-glue
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill uses a helper script
scripts/atoms-search.pyto retrieve knowledge snippets (referred to as 'atoms') from the vendor's official domain athttps://api.xyskill.xyz. This is used to provide evidence-based recommendations during the analysis phase. - [COMMAND_EXECUTION]: The skill instructions direct the agent to execute a local Python utility,
scripts/atoms-search.py, to facilitate keyword-based lookups within both local data files and the remote vendor API. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process user-provided transcripts to check for 'watery' content or logical breaks. It uses content from these transcripts to generate search queries for its internal tools.
- Ingestion points: User-provided video transcripts submitted via the agent interface (Phase 1).
- Boundary markers: The skill employs a structured phase-based execution flow to maintain context, though it does not explicitly define escape sequences for transcript content.
- Capability inventory: The agent can execute the bundled
atoms-search.pyscript via a shell command (python3). - Sanitization: The skill provides specific instructions to the agent on how to construct search keywords (limiting length and complexity) but does not explicitly mandate shell-safe escaping for those terms.
Audit Metadata