skills/wso2/agent-skills/api-publish/Gen Agent Trust Hub

api-publish

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTIONPERSISTENCEPRIVILEGE_ESCALATION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill downloads the ap CLI binary and the API Gateway release from the official WSO2 GitHub repository during the setup phase.
  • [COMMAND_EXECUTION]: Executes shell commands and PowerShell scripts to identify host IP addresses for Docker networking, manage Docker Compose stacks, and interact with the ap CLI.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external data in the form of OpenAPI specifications provided by the user to generate gateway deployment configurations.
  • Ingestion points: Phase 2 of SKILL.md ingests OpenAPI specs provided as file paths or raw text.
  • Boundary markers: None explicitly implemented for the ingested spec content.
  • Capability inventory: The skill uses ap gateway apply (writes to the gateway), node (executes bundled scripts), and curl (network testing).
  • Sanitization: No specific sanitization or validation of the OpenAPI content is mentioned before processing.
  • [DYNAMIC_EXECUTION]: Uses Node.js to execute bundled JavaScript files for installation (install-ap-cli.js), environment setup (setup-gateway.js), and configuration management (init-local-cli-config.js).
  • [PRIVILEGE_ESCALATION]: Modifies the user's environment by appending the ap CLI installation path to shell profile files (.bashrc, .zshrc) or the Windows User PATH to ensure the tool is available in future terminal sessions.
  • [PERSISTENCE]: Modifies shell startup scripts to ensure the ap command remains accessible across different sessions, which is standard behavior for tool installers.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 12:05 PM