The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes data from locally stored markdown files to track user preferences, trade journals, and learning progress. This creates a surface where the agent could potentially ingest instructions embedded within these files.
Ingestion points: ~/Clawic/data/trading/memory.md, ~/Clawic/data/trading/journal.md, and ~/Clawic/data/trading/progress.md.
Boundary markers: None identified in the provided file templates to separate user data from agent instructions.
Capability inventory: The skill is restricted to local file operations within its designated data directory. It does not possess network access or arbitrary command execution capabilities.
Sanitization: No explicit sanitization or filtering of the content read from the memory files is mentioned.
[COMMAND_EXECUTION]: The skill provides documentation for initial setup using shell commands to create the necessary directory structure and initialize files.
Evidence: mkdir -p ~/trading and touch ~/Clawic/data/trading/{memory.md,journal.md,progress.md} in memory-template.md and setup.md.
Context: These are benign, standard file system operations intended for environment initialization and do not represent a security risk.

Trading