The Agent Skills Directory

[COMMAND_EXECUTION]: The skill includes a shell script (evaluate-session.sh) intended to run as a session-end hook. The script performs environment checks, directory management (mkdir -p), and transcript analysis using grep and jq.- [DATA_EXPOSURE]: The skill reads session transcripts to determine if extraction thresholds are met. This involves local file access to data that may contain sensitive user interactions, which is required for the skill's primary function.- [INDIRECT_PROMPT_INJECTION]: The skill evaluates session logs containing untrusted data from user prompts.
Ingestion points: The evaluate-session.sh script parses the session transcript file provided at runtime.
Boundary markers: No specific boundary markers or instruction-ignoring warnings are implemented in the script logic that counts messages.
Capability inventory: The skill possesses file system write capabilities within the user's home directory (~/.claude/skills/learned/) and configuration of shell-based hooks.
Sanitization: The script utilizes jq for structured data parsing where available, which is a robust practice against malformed input.

continuous-learning