The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill instructs the agent to prompt the user to install a third-party GitHub CLI extension from an unverified community source.
Evidence: The workflow includes the command gh extension install yahsan2/gh-sub-issue in SKILL.md. This extension executes code with the privileges of the authenticated GitHub user.
[INDIRECT_PROMPT_INJECTION]: The skill ingests data from local repository files and interpolates the content into GitHub issues without sanitization, creating a surface for indirect prompt injection if the repository contains malicious instructions.
Ingestion points: The skill reads CLAUDE.md, AGENTS.md, and CONTRIBUTING.md to extract conventions and coding standards.
Boundary markers: No explicit delimiters or instructions are provided to the LLM to ignore potentially malicious embedded content within these files.
Capability inventory: The skill uses gh issue create, gh issue edit, and gh project item-edit to write data back to GitHub.
Sanitization: There is no evidence of validation or filtering for the content extracted from the repository files before it is included in the generated issue bodies.
[COMMAND_EXECUTION]: The skill performs extensive shell command execution using the GitHub CLI (gh) and jq to manage repository metadata and project boards.
Evidence: Numerous bash code blocks are defined for operations such as listing collaborators, creating labels, and editing project items based on dynamically discovered field IDs.

github-project-manager