The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the GitHub CLI (gh) to perform repository analysis and file tree retrieval. It also uses a local CLI tool (qmd) to index generated brief files. These operations are functional requirements for the skill's primary purpose of library scoping and information retrieval.
[PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection due to its core function of analyzing external repository data. 1. Ingestion points: Repository structure listings and configuration files (e.g., package.json, pyproject.toml) are retrieved via the gh tool in steps-c/step-02-analyze-target.md. 2. Boundary markers: No explicit delimiters or 'ignore embedded instructions' warnings are implemented to wrap the untrusted repository data during analysis. 3. Capability inventory: The skill can write YAML files to the local directory and execute CLI commands (gh, qmd) in steps-c/step-05-write-brief.md. 4. Sanitization: Output filenames and identifiers are restricted to kebab-case and critical decisions require interactive user confirmation, which provides a layer of defense against certain injection-based file system or command abuses.

skf-brief-skill