The Agent Skills Directory

[COMMAND_EXECUTION]: The skill invokes local system utilities including git, gh (GitHub CLI), and bun to extract project metrics. Specifically, it executes a local TypeScript script (fleet-scan.ts) to generate a 'family tree' of the project's instances.
[DATA_EXFILTRATION]: Accesses the local file system to read project files such as src/cli/index.ts and inspects the directory structure in $HOME/Code/github.com/Soul-Brews-Studio/ to retrieve Git history and repository counts.
[EXTERNAL_DOWNLOADS]: Communicates with GitHub servers using the gh tool to list repositories and fetch issue content specifically from the Soul-Brews-Studio organization.
[PROMPT_INJECTION]: The skill ingests external content from GitHub issues, which represents a potential surface for indirect prompt injection.
Ingestion points: External data is retrieved via gh issue view 60 --repo Soul-Brews-Studio/arra-oracle-v3.
Boundary markers: No explicit delimiters or instructions are provided to the agent to ignore potentially malicious content within the issue body.
Capability inventory: The skill has the ability to execute local scripts via bun, run shell commands, and read local files.
Sanitization: The skill uses basic filtering (via grep and head) but does not perform comprehensive sanitization or validation of the fetched external text.

about-oracle