The Agent Skills Directory

[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection (Category 8) due to the ingestion and processing of untrusted content.
Ingestion points: SKILL.md defines input handling that includes reading local files and fetching content from arbitrary URLs via WebFetch.
Boundary markers: The 'Scoring Process' does not define specific delimiters (e.g., XML tags or triple quotes) or provide system-level instructions for the agent to ignore instructions embedded within the blog post.
Capability inventory: The skill possesses file system read access and network ingress capabilities via WebFetch.
Sanitization: There is no mention of input validation, sanitization, or escaping of the retrieved content before it is interpolated into the analysis prompt.
[EXTERNAL_DOWNLOADS]: The skill uses the WebFetch tool to download content from external, user-specified URLs as part of its primary blog auditing function.

blog-analyze