skills/jimweller/clanker-skills/ado/Gen Agent Trust Hub

ado

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPRIVILEGE_ESCALATIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill performs extensive infrastructure management tasks by executing az CLI and curl commands to interact with Azure DevOps repositories, pipelines, and agent pools.
  • [PRIVILEGE_ESCALATION]:
  • Includes functionality to bypass established repository branch policies using the --bypass-policy true flag during pull request completion.
  • Contains commands to grant service connections and agent queues broad authorization for all pipelines within a project.
  • [DATA_EXFILTRATION]: Provides templates for creating service hook subscriptions (webhooks) that can be configured to send event data to external, user-specified URLs.
  • [INDIRECT_PROMPT_INJECTION]:
  • Ingestion points: The skill ingests data from external sources that may be controlled by third parties, including pull request comments, build logs, and code search results via REST API endpoints.
  • Boundary markers: No specific delimiters or instructions to ignore embedded commands are used when processing external data.
  • Capability inventory: The skill possesses high-privilege capabilities including repository deletion (az repos delete), pipeline deletion (az pipelines delete), and the ability to modify repository content via the git push REST API.
  • Sanitization: There is no evidence of input validation or content sanitization applied to the data retrieved from external Azure DevOps resources before it is processed or returned to the agent context.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 06:30 PM
Security Audit — agent-trust-hub — ado