runway-dev-workflows
Pass
Audited by Gen Agent Trust Hub on Aug 30, 2026
Risk Level: SAFECREDENTIALS_UNSAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [CREDENTIALS_UNSAFE]: The skill instructs the agent to probe the workspace for the
RUNWAYML_API_SECRETenvironment variable. Although it explicitly directs the agent not to print the secret, accessing API credentials from the environment is a sensitive operation that carries an inherent risk of data exposure. - [EXTERNAL_DOWNLOADS]: The skill references and fetches developer guidance from
https://docs.dev.runwayml.com/llms.txt. This is a vendor-owned resource used to provide up-to-date integration instructions. - [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external URLs and the output of the
get_workflows_setup_guideMCP tool, which constitutes a surface for potential instruction injection. - Ingestion points: Technical documentation hosted at
docs.dev.runwayml.comand results from setup guide tool calls. - Boundary markers: The skill does not define specific delimiters or instructions to the agent to treat the fetched content as untrusted or to ignore any commands embedded within the data.
- Capability inventory: The agent is authorized to build, modify, and verify API integrations, including running workflow invocations.
- Sanitization: No explicit sanitization or validation of the documentation content is performed before the agent processes it.
Audit Metadata