The Agent Skills Directory

[COMMAND_EXECUTION]: The skill includes instructions to execute a local Python script for workspace initialization.
Execution method: The documentation in SKILL.md directs the agent to run scripts/run.py using the uv package manager.
Evidence: uv run python .codex/skills/thesis-question-list/scripts/run.py --workspace <workspace>.
Context: The script is a simple utility that creates a markdown skeleton (question_list.md) if it does not exist; it does not utilize eval() or process arbitrary shell commands.
[INDIRECT_PROMPT_INJECTION]: The skill acts as a data aggregator, processing various thesis-related documents which introduces an indirect prompt injection surface.
Ingestion points: As listed in SKILL.md, the skill reads content from multiple external files including codex_md/material_index.md, codex_md/missing_info.md, and claude_md/review_checklist.md.
Boundary markers: The skill does not define specific delimiters or "ignore embedded instructions" warnings for the processed data.
Capability inventory: The skill possesses file writing capabilities via scripts/run.py to update the thesis control plane.
Sanitization: The instructions do not specify any sanitization or validation logic for the content ingested from the external markdown files before processing.

thesis-question-list