The Agent Skills Directory

[COMMAND_EXECUTION]: In scripts/create_issue.py, the script executes os.system(f'python {script}') to trigger metadata extraction. The script path, derived from the file system, is not quoted. If the skill is installed in a directory containing spaces or shell metacharacters (e.g., ;, &), it could lead to arbitrary command execution.
[EXTERNAL_DOWNLOADS]: The scripts/extract_issue_metadata.py file imports the yaml (PyYAML) library, but the skill does not provide a requirements.txt or similar manifest to declare this dependency or ensure a verified version is used.
[PROMPT_INJECTION]: The skill implements an indirect prompt injection surface.
Ingestion points: Reads all .md files in .docs/issues/ or _docs/issues/ to extract metadata.
Boundary markers: None; the script parses YAML frontmatter and regex-matched text directly from user-controlled or external files.
Capability inventory: The skill has file-write capabilities (generating index.md).
Sanitization: There is no sanitization of the metadata extracted (e.g., title, author, status) before it is written into the index.md summary table. If an attacker places a malicious markdown file with instructions in the metadata fields, an agent reading the resulting index could be manipulated.

md-issue-writer