docyrus-cli-app
Pass
Audited by Gen Agent Trust Hub on Apr 11, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill facilitates the execution of the docyr us CLI tool for platform operations including authentication, data record management (ds), and developer studio schema operations (studio).
- [EXTERNAL_DOWNLOADS]: Dow n load s tenant OpenAPI specifications via the docyr us discover api command and requires the local presence of the Bun run time to execute the terminal UI (docyr us tui).
- [DATA_EXFILTRATION]: Provides the ability to send arbitrary data and HTTP requests to external en d point s through the docyr us curl and docyr us discover con n ector-curl command s, including the use of stored con n ector authentication to interact with third-party provider s.
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its core function of processing untrusted external data. \n * Ingestion point s: Data enter s the agent's context through API respon ses from docyr us ds list, OpenAPI schema s retrieved via docyr us discover, and result s from external con n ector action s. \n * Boun d ary marker s: There are no defined delim iter s or instruction s to the agent to disregar d natural language instruction s that might be em bed ded in the retrieved JSON or YAML data. \n * Capability inventor y: The agent has acces s to network operation s (curl), local file reading (--from-file), and interactive shell execution (tui). \n * Sanitization: The skill does not specify any validation or sanitization logic for content fetched from external API en d point s before it is processed by the agent.
Audit Metadata