The Agent Skills Directory

[SAFE]: No security issues were detected. The skill provides instructions for a post-deployment monitoring workflow, including HTTP status checks, performance metrics, and local logging, which are consistent with its stated purpose.
[PROMPT_INJECTION]: The skill processes content from external URLs (HTML, console logs, API responses), which creates an inherent surface for indirect prompt injection. This is a functional requirement for its monitoring purpose.
Ingestion points: Content, console logs, and API response data from monitored URLs as described in SKILL.md.
Boundary markers: No specific delimiters or instructions to ignore embedded commands are defined in the instructions.
Capability inventory: Local file logging to ~/.claude/canary-watch.log and network operations for monitoring and optional webhooks.
Sanitization: No explicit sanitization or validation of remote content is described in the provided text.

canary-watch