The Agent Skills Directory

[COMMAND_EXECUTION]: The 'Memory Protocol' in SKILL.md instructs the agent to execute the shell command cat .claude/context/memory/learnings.md at the start of the session to retrieve state.
[DATA_EXFILTRATION]: The skill is configured to access .claude/context/memory/learnings.md, which is a file path used for internal agent context and memory management. Accessing internal system or application state files increases the potential for data exposure.
[PROMPT_INJECTION]: The skill consumes data from an external file (learnings.md) and incorporates it into the active context without safety boundaries.
Ingestion points: The file .claude/context/memory/learnings.md is read via a shell command defined in SKILL.md.
Boundary markers: The instructions lack delimiters or warnings to treat the ingested memory content as potentially untrusted data.
Capability inventory: The skill is granted Read, Write, and Edit tool permissions, along with the ability to execute shell commands.
Sanitization: No validation or sanitization is performed on the content retrieved from the memory file before it is processed by the agent.

tauri-svelte-ui-components