The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill's primary purpose is to consolidate and organize other skill files (SKILL.md). This creates a vulnerability surface where the agent might inadvertently execute instructions contained within the skills it is currently processing.
Ingestion points: The skill discovers and reads SKILL.md files from multiple source directories and GitHub repositories during the 'Discovery & Analysis' phase.
Boundary markers: No specific delimiters or instructions to ignore embedded commands are present in the consolidation workflow.
Capability inventory: The skill utilizes directory manipulation, file writing, and tool integrations such as git-workflow-helper and github-auth to modify local files and push to remote repositories.
Sanitization: While the documentation mentions an optional scan_for_secrets check, there is no explicit sanitization or filtering for natural language instructions that could trigger unintended agent behavior.
[EXTERNAL_DOWNLOADS]: The consolidation workflow involves fetching data from remote sources, specifically GitHub repositories, using the bulk-github-skills-downloader integration.
[COMMAND_EXECUTION]: The skill performs extensive file system operations, including directory flattening, file renaming, and repository maintenance tasks which involve executing commands through the CLI environment.

skills-consolidator