The Agent Skills Directory

[PROMPT_INJECTION]: The skill analyzes existing project content to determine the current state of error monitoring, which introduces a surface for indirect prompt injection.
Ingestion points: Project source files (.swift) and configuration files (Package.swift) are read via Bash commands defined in SKILL.md.
Boundary markers: Absent. No explicit delimiters are used to wrap or isolate the ingested project content.
Capability inventory: The skill uses Bash, Write, and Edit tools to perform project modifications based on the context found.
Sanitization: Absent. The content of the read files is processed by the agent without explicit filtering or sanitization.
[COMMAND_EXECUTION]: The skill uses the Bash tool to execute rg (ripgrep) and cat for project inspection. These commands are employed for the legitimate purpose of identifying existing crash reporting SDKs and error-handling patterns within the codebase.
[EXTERNAL_DOWNLOADS]: The skill facilitates the integration of official SDKs and configuration instructions from well-known services, including Sentry (github.com/getsentry/sentry-cocoa) and Firebase Crashlytics. The skill provides templates and documentation for these reputable industry-standard providers.

error-monitoring