The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses subprocess.run in scripts/gen_repo_catalog.py to execute git commands. Specifically, it uses git clone to download repositories, git remote get-url to identify the repository source, and git log to retrieve the latest commit date. These commands are executed with controlled arguments and are central to the skill's stated purpose of indexing code repositories.
[EXTERNAL_DOWNLOADS]: The skill fetches external content by cloning GitHub repositories to a user-defined "upstream" directory. This involves downloading third-party code from remote servers. The skill explicitly handles cloning failures (e.g., due to network issues or private repositories) by falling back to a minimal metadata-only entry.
[INDIRECT_PROMPT_INJECTION]: The skill ingests data from external README.md files to generate repository summaries.
Ingestion points: The first_summary_line function in scripts/gen_repo_catalog.py reads the content of README files from the cloned repositories.
Boundary markers: The extracted summary is placed directly into the Markdown body of the project card without specific boundary markers or warnings about external content.
Capability inventory: The skill has the capability to execute shell commands (git) and write files to the local file system within the Obsidian vault and the upstream cache.
Sanitization: The skill includes a clean_line function that uses regular expressions to strip HTML tags, Markdown images, links, and formatting markers from the extracted text to ensure the resulting summary is plain text.

soia-pkm-clip-github-repo