deploy-to-hosting
Pass
Audited by Gen Agent Trust Hub on Oct 28, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides instructions for the agent to use local shell utilities such as
curl,git, andzipto archive project files and upload them to the hosting provider. - [DATA_EXFILTRATION]: The instructions direct the agent to read sensitive local files, specifically
.envfiles containing environment variables, to sync them with the remote hosting environment. Data is transmitted to the vendor's infrastructure using secure upload URLs generated via the MCP server. - [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes untrusted local project data, which constitutes an attack surface for indirect prompt injection.
- Ingestion points: The agent reads local project files, directory structures, and environment configurations from the disk (SKILL.md).
- Boundary markers: The instructions do not specify the use of delimiters or ignore-instructions for the contents of the project files being processed.
- Capability inventory: The skill utilizes file system read access, network transmission via
curl, and source control management viagit(SKILL.md). - Sanitization: No explicit sanitization or validation of the ingested code/data content is performed prior to deployment processing.
Audit Metadata