spec-from-screenshot
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data in the form of user-provided screenshots.
- Ingestion points: The workflow defined in
SKILL.md(Step 1 and Step 2) requires the agent to ingest and analyze screenshots provided by the user in the conversation context. - Boundary markers: The instructions lack explicit boundary markers or directives for the agent to ignore or isolate text content found within images that might be interpreted as system instructions.
- Capability inventory: The skill utilizes the capability to write a file (
SPEC.md) to the local file system as detailed in Step 3. - Sanitization: There is no evidence of sanitization, validation, or filtering of the content extracted from screenshots before it is interpolated into the generated documentation.
Audit Metadata