travel-city-walk-by-1uczxsi
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references external URLs in
references/cases.md(specifically goodcase.ai) and instructs the agent to access them using browser or media tools to analyze the content. While the links appear relevant to the skill's purpose, they point to non-whitelisted domains, which introduces a surface for network communication with external entities.\n- [INDIRECT_PROMPT_INJECTION]: The skill's workflow requires the agent to ingest and analyze data from external websites, creating a potential vector for indirect prompt injection.\n - Ingestion points: The agent is instructed in
SKILL.mdto "Inspect the anchor's finished media with an available image, video, or browser tool." The targets for this inspection are the external URLs listed inreferences/cases.md.\n - Boundary markers: Absent. The skill does not provide instructions to the agent to treat the external content as untrusted or to ignore any potential instructions embedded within the media or the hosting pages.\n
- Capability inventory: The skill uses text generation capabilities to produce detailed video prompts, shot-by-shot motion briefs, and reference contracts. While it does not have file-write or command-execution capabilities, the generated prompts could be manipulated by external content.\n
- Sanitization: Absent. The skill does not specify any sanitization, filtering, or validation steps for the data retrieved from external sources before it is used to influence the agent's output.
Audit Metadata