The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local scripts validate-yaml.py and search-yaml.py to verify the integrity of the manifest and search for outdated documentation entries.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted data from source code files to generate documentation.
Ingestion points: Reads source code from paths defined in source_root and source_files (identified in SKILL.md and reference.md).
Boundary markers: Absent. There are no instructions or delimiters specified to prevent the agent from obeying instructions hidden within the source code being analyzed.
Capability inventory: The agent has the ability to read arbitrary files, write to the docs/api/ directory, and execute shell commands (via the Python validation scripts).
Sanitization: Absent. The instructions do not define any filtering or sanitization of comments, strings, or metadata extracted from the source files before they are incorporated into the output or agent context.

cs-libdoc