skills/pkulik0/gitea-skill/gitea/Gen Agent Trust Hub

gitea

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill makes extensive use of the tea CLI via bash to perform Gitea operations. This includes powerful capabilities such as user administration (tea admin users), SSH key management (tea ssh-keys), and repository migration, all of which are documented as intended functionality for managing a Gitea instance.
  • [INDIRECT_PROMPT_INJECTION]: The skill reads untrusted content from external Gitea instances, including issue descriptions, pull request details, and comments. This presents an attack surface where a repository's content could attempt to influence the agent's behavior.
  • Ingestion points: Gitea issue/PR listings and detail views accessed via tea issues ls, tea pulls ls, and tea comments ls.
  • Boundary markers: None explicitly defined in the provided instruction set.
  • Capability inventory: Access to all tea CLI subcommands, shell execution, and the tea api for arbitrary requests.
  • Sanitization: Relies on the tea CLI's internal argument handling and standard shell quoting practices recommended in the instructions.
  • [EXTERNAL_DOWNLOADS]: The skill references the official Gitea CLI (tea) and its repository on gitea.com. As Gitea is a well-known service and the source is official, these references are considered safe and consistent with the skill's purpose.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 05:16 PM