The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill creates an ingestion surface by scanning and analyzing the text contents of various project files to estimate token consumption.
Ingestion points: Processes content from local files including agents/*.md, skills/*/SKILL.md, rules/**/*.md, rules/**/*.mdc, and CLAUDE.md during the Inventory phase.
Boundary markers: No explicit delimiters or instructions to ignore embedded commands are specified for the agent when reading these files.
Capability inventory: The skill identifies as a meta-analysis tool and does not possess capabilities for command execution, network requests, or file writing.
Sanitization: No sanitization or filtering of the ingested file content is mentioned prior to token estimation.
[SAFE]: The skill's functionality is purely analytical and restricted to local file metadata and content counting. It does not exhibit patterns of data exfiltration, persistence, or privilege escalation.

context-budget