The Agent Skills Directory

[PERSISTENCE]: The skill instructs the agent to use CronCreate or the schedule skill to create one-time tasks for verifying fixes after a deployment. While this is a documented feature for task automation, it constitutes a persistence mechanism that schedules future automated execution.
[TIME_DELAYED_CONDITIONAL]: The verification logic is designed to trigger after a 24–48 hour delay, gating specific diagnostic and reporting operations behind a time-based condition.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external sources, specifically stack traces and error messages from the Nightwatch MCP server, which could contain malicious instructions designed to influence agent behavior.
Ingestion points: Data retrieved via get_issue and list_issues tools (SKILL.md).
Boundary markers: Absent. The instructions do not define delimiters or warnings to separate external issue content from agent instructions.
Capability inventory: The skill possesses capabilities to modify application code, update container and infrastructure configurations, and perform network operations via the Nightwatch API (update_issue, add_issue_comment).
Sanitization: Absent. There is no mention of sanitizing, escaping, or filtering the retrieved issue data before the agent processes it.

fix-nightwatch-issues