The Agent Skills Directory

[SAFE]: The skill consists entirely of markdown instructions for planning context allocation and cache optimization. It does not include any scripts, commands, or automation that could be exploited.
[SAFE]: References to external components (e.g., token-cost, repo-map, context-crusher) are relative links to other local markdown files within the same environment, representing an integrated documentation structure rather than remote dependencies.
[SAFE]: There are no indicators of prompt injection, obfuscation, or data exfiltration. The instructions focus on architectural best practices for token management and cache-aware prefixing.

context-budget