The Agent Skills Directory

[SAFE]: The skill's primary function is to audit a project stack for cost comparison purposes. It focuses on identifying service providers through dependency manifests, configuration files (e.g., vercel.json, netlify.toml), and environment variable keys (e.g., SENTRY_DSN, DATABASE_URL).
[CREDENTIALS_SAFE]: The skill includes explicit instructions to avoid reading or printing secret values. It mandates the use of commands like cut -d= -f1 when inspecting environment files to ensure only keys, and never values, are processed or reported.
[EXTERNAL_DOWNLOADS]: No external code downloads or remote script executions were detected. All reference data for pricing is bundled locally within the skill in the references/pricing.md file.
[COMMAND_EXECUTION]: The skill uses standard local file system inspection and grep-like patterns to identify infrastructure components. It does not invoke dangerous system commands, privilege escalation, or persistence mechanisms.
[INDIRECT_PROMPT_INJECTION]: While the skill processes project data, it focuses on structured configuration and metadata. The instructions provide clear boundaries for how to handle this data, and the final report follows a strict markdown template which limits the risk of processed data influencing the agent's behavior maliciously.

estimate-temps-savings