The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from user-supplied materials and existing website content to develop marketing briefs. 1. Ingestion points: SKILL.md instructions to read 'supplied material' and 'inspect relevant existing content'. 2. Boundary markers: Absent; the skill does not define specific delimiters or instructions to ignore embedded commands in external data. 3. Capability inventory: The skill is restricted to reading and writing markdown documentation files within the docs/webcraft/ project directory. 4. Sanitization: Absent; no explicit sanitization or filtering logic is defined for the ingested material.
[SAFE]: The skill includes explicit instructions in references/artifact.md to keep credentials and private analytics exports out of the record, which aligns with security best practices for credential management.

webcraft-brief