The Agent Skills Directory

[COMMAND_EXECUTION]: The skill includes a pre-tool hook script (hooks/pre-tool.sh) and a search utility (scripts/search_lessons.py) that are executed during normal operation to monitor tool usage and query lesson databases.
[REMOTE_CODE_EXECUTION]: Instructions in SKILL.md reference the use of python scripts/install-external-skills.py for adding external capabilities. This script is missing from the provided skill files, representing an unverifiable code execution dependency.
[EXTERNAL_DOWNLOADS]: The skill references external skill registries and sources through relative paths (../../../references/external-skills.yaml), indicating reliance on resources stored outside the skill's own directory scope.
[PROMPT_INJECTION]: The skill's architecture is vulnerable to indirect prompt injection because it ingests and processes lesson content from markdown files that may be generated from external or untrusted user corrections.
Ingestion points: scripts/search_lessons.py recursively reads and extracts text from all markdown files in the directory.
Boundary markers: The skill does not employ delimiters or specific instructions to prevent the agent from following commands embedded within the lesson data.
Capability inventory: The agent possesses script execution capabilities via the included bash and Python scripts.
Sanitization: There is no evidence of content sanitization or validation performed on the lesson data extracted from the markdown files.

lessons