The Agent Skills Directory

[COMMAND_EXECUTION]: The skill includes a Python script scripts/slo-burn-rate.py which is a pure utility for calculating error budgets. It uses only standard libraries (argparse, json, sys, math) and performs no risky operations such as shell execution, file system modification, or network access.
[DATA_EXPOSURE]: The templates provided (e.g., templates/runbook-template.md, templates/error-budget-policy.md) contain standard placeholders for service information such as [YOUR_API_KEY_HERE] or [redacted]. No actual sensitive data or hardcoded credentials were found.
[REMOTE_CODE_EXECUTION]: There are no patterns of remote code execution. All logic is contained within the local files, and no external dependencies are downloaded or executed at runtime.
[INDIRECT_PROMPT_INJECTION]: The skill defines a surface for processing SLI data through its calculation script. However, the script strictly parses structured JSON and performs mathematical operations without treating inputs as instructions or executable code.
[DYNAMIC_CONTEXT_INJECTION]: The SKILL.md file was scanned for pre-execution shell command syntax (!command). While it contains command examples for third-party incident tools (e.g., !ic declare), these are documented as text examples and do not utilize the platform's dynamic execution syntax.

site-reliability-engineering