The Agent Skills Directory

[COMMAND_EXECUTION]: The skill employs standard shell utilities including head, grep, echo, dirname, and basename to read and display headers of files located in ~/.claude/skills/, ~/.claude/agents/, and ~/.claude/commands/. These operations are used purely for local analysis and discovery of existing artifacts.
[PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection because it reads and processes instructions from existing local files that could contain malicious payloads.
Ingestion points: Content is ingested from local artifact files via SKILL.md (Step 2: Search Existing Artifacts).
Boundary markers: Absent; the skill does not use specific delimiters or warnings to ignore embedded instructions within the analyzed files.
Capability inventory: The skill uses shell execution capabilities to read files and perform keyword searches.
Sanitization: There is no evidence of sanitization or filtering applied to the content retrieved from existing artifacts before it is processed by the agent.

creation-guard