cloud
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONDATA_EXFILTRATIONCREDENTIALS_UNSAFEEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes test cases, templates, and TypeScript functions retrieved from the Shiplight cloud API or local repositories.\n
- Ingestion points: External content is ingested via MCP tools (e.g.,
get_test_case) and REST API endpoints (e.g.,/v1/test-cases).\n - Capability inventory: The agent is authorized to execute shell commands (
curl), modify local configuration files (e.g.,.env), and write data to the cloud API.\n - Boundary markers: The instructions do not define specific delimiters or instructions to ignore embedded commands in the processed data.\n
- Sanitization: There is no mention of sanitization or validation of the fetched YAML or code content.\n- [COMMAND_EXECUTION]: The skill uses
curlto interact with the Shiplight REST API and provides examples for downloading binary files to the local file system.\n- [DATA_EXFILTRATION]: Facilitates the synchronization of local project files and sensitive browser session states (storage_state_path) to the vendor's cloud infrastructure.\n- [CREDENTIALS_UNSAFE]: The skill manages sensitive information, including API tokens and test account credentials. It provides mechanisms for storing tokens in.envfiles and uploading local authentication sessions to the cloud.\n- [EXTERNAL_DOWNLOADS]: Fetches test data and artifacts from Shiplight's official API and S3-backed storage.
Audit Metadata