The Agent Skills Directory

[SAFE]: The skill consists exclusively of instructional Markdown files that provide templates and personas for technical writing tasks. It does not perform any network operations, access sensitive local files, or include hidden instructions.\n- [NO_CODE]: No executable scripts, binaries, or configuration files that install dependencies are present in the skill package. All functionality is driven by natural language prompting.\n- [INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection as it is designed to ingest and process untrusted external data such as code and git diffs.\n
Ingestion points: External code snippets, git diffs, and technical requirements are ingested in explain_code.md, summarize_git_diff.md, and write_pull_request.md.\n
Boundary markers: The instructions lack explicit delimiters (e.g., XML tags) to separate user-provided data from agent instructions, relying on the agent's context window.\n
Capability inventory: The skill uses standard agent capabilities for text analysis; no specific tools for file writing, network access, or command execution are invoked by the instructions.\n
Sanitization: No input sanitization, filtering, or validation is performed on the ingested content.

technical-documentation