The Agent Skills Directory

[DATA_EXFILTRATION]: The skill reads application-specific session metadata from ~/.claude/sessions/. This directory contains session history and state for the Claude Code tool. Accessing hidden application directories in the home folder exposes internal configuration and usage patterns.
[COMMAND_EXECUTION]: The skill uses the Bash tool to execute file system commands including cat and ls on sensitive application paths and project directories to extract status information.
[INDIRECT_PROMPT_INJECTION]: The skill processes content from local files that may contain untrusted data, creating a surface for indirect prompt injection.
Ingestion points: Reads JSON data from ~/.claude/sessions/*/context-monitor-cache.json and Markdown content from quality_reports/plans/ and quality_reports/session_logs/.
Boundary markers: Absent. The skill does not use specific delimiters or instructions to treat the ingested file content as untrusted data.
Capability inventory: The skill utilizes Bash, Read, and Glob tools to retrieve and process information.
Sanitization: No sanitization, escaping, or schema validation is performed on the file content before it is processed by the agent.

context-status