The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands with user-provided arguments. Evidence: python3 bin/semantic-search.py "your question" in SKILL.md. Risk: Direct interpolation of user input into shell commands can lead to command injection if the underlying execution environment does not perform proper sanitization.
[COMMAND_EXECUTION]: The skill includes instructions to execute maintenance scripts. Evidence: python3 bin/semantic-index.py --rebuild in SKILL.md.
[PROMPT_INJECTION]: The skill contains an indirect prompt injection surface. 1. Ingestion points: Output from semantic-search.py and contents of vault notes (SKILL.md). 2. Boundary markers: Absent; there are no instructions to use delimiters or ignore instructions embedded within the retrieved data. 3. Capability inventory: Shell execution capabilities via python3 (SKILL.md). 4. Sanitization: Absent; the instructions do not require the agent to validate or filter the content retrieved from the vault.

retrieval