The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill uses npx to download and execute the ruv-swarm and claude-flow@alpha packages from the npm registry during execution.
[REMOTE_CODE_EXECUTION]: Executes external code downloaded via npx for repository management tasks and references the third-party GitHub Action ruvnet/swarm-action@v1 in its documentation.
[COMMAND_EXECUTION]: Extensively utilizes the GitHub CLI (gh) to read and write issue data, manage project boards, and create items, providing a broad capability set for repository manipulation.
[PROMPT_INJECTION]: An indirect prompt injection surface is created by the skill's ingestion of untrusted data from GitHub issues to drive automated decisions and actions.
Ingestion points: The skill reads issue bodies and comments via gh issue view in references/command-reference.md.
Boundary markers: Instructions do not include explicit delimiters or safety warnings to ignore instructions embedded within the processed issue data.
Capability inventory: The skill possesses write access to repository metadata and issues, and can execute external shell commands.
Sanitization: There is no evidence of sanitization or validation of the untrusted issue content before it is used in subsequent operations or process generation.

github-project-management