copy-adapt
Pass
Audited by Gen Agent Trust Hub on Nov 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests 'Source copy' provided by the user in Step 1 to perform transformation tasks. This represents a potential surface where untrusted data could contain instructions meant to influence the agent's logic during the adaptation process.
- Ingestion points: 'Source copy' input in SKILL.md (Step 1).
- Boundary markers: The skill does not define specific delimiters or warnings to ignore instructions embedded within the source copy.
- Capability inventory: The skill's capabilities are limited to text analysis, routing to other internal domain skills, and loading a local reference file (
anti-slop.md). It lacks dangerous capabilities such as network access, file-writing, or shell command execution. - Sanitization: There is no evidence of sanitization or filtering applied to the source text before it is processed.
Audit Metadata