The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes untrusted content from external issue trackers (e.g., GitHub issue bodies) or local markdown files to drive autonomous code implementation, which creates a surface for Indirect Prompt Injection.\n
Ingestion points: The skill reads parent issues, linked child tickets, and backlog files in the .scratch/<feature>/issues/ directory via the gh CLI or local file system.\n
Boundary markers: The instructions do not define delimiters or explicit 'ignore instructions' warnings when passing external ticket content to implementation sub-agents.\n
Capability inventory: Spawns agents that can perform file edits, run shell commands (git, package managers, test runners), and post status updates via the gh CLI.\n
Sanitization: No input sanitization or validation of the ingested ticket content is described in the skill instructions.\n- [EXTERNAL_DOWNLOADS]: The skill declares a dependency on the external repository mattpocock/skills in the YAML frontmatter and references a setup command (/setup-matt-pocock-skills) to install external components.\n- [COMMAND_EXECUTION]: To operate autonomously, the skill requires the user to allow-list several shell commands, including git, gh (GitHub CLI), and various project-specific tools like test runners and package managers, which are then executed without human confirmation.

flock-agents