The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill references an external MCP server endpoint at https://rube.app/mcp to provide the underlying GitHub tool functionality.
[COMMAND_EXECUTION]: The skill provides instructions for executing high-privilege GitHub operations, such as merging pull requests (GITHUB_MERGE_A_PULL_REQUEST), deleting repositories (GITHUB_DELETE_A_REPOSITORY), and modifying branch protection rules (GITHUB_UPDATE_BRANCH_PROTECTION). It mitigates risk by instructing the agent to require explicit human confirmation for these actions.
[DATA_EXFILTRATION]: The skill is designed to read potentially sensitive repository data, including source code, private issue details, and commit history, as part of its intended automation functionality.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes untrusted data from GitHub repositories while maintaining write access to the environment.
Ingestion points: The agent reads data through GITHUB_LIST_REPOSITORY_ISSUES, GITHUB_GET_A_PULL_REQUEST, and GITHUB_SEARCH_CODE (all in SKILL.md).
Boundary markers: No explicit delimiters or instructions are provided to help the agent distinguish between data and embedded instructions in the fetched content.
Capability inventory: The skill includes write/admin capabilities such as GITHUB_MERGE_A_PULL_REQUEST and GITHUB_CREATE_A_WORKFLOW_DISPATCH_EVENT (all in SKILL.md).
Sanitization: There are no specified sanitization or validation routines for the content retrieved from GitHub before it is processed by the agent.

github-automation