The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute the claude command-line utility with background flags (--bg) to spawn new processes.
[INDIRECT_PROMPT_INJECTION]: 1. Ingestion points: The skill summarizes the current conversation history as described in SKILL.md. 2. Boundary markers: There are no specific delimiters or instruction-isolation markers defined for the handoff summary content. 3. Capability inventory: The skill can execute shell commands via the claude CLI as documented in SKILL.md. 4. Sanitization: The instructions include a manual redaction policy for sensitive data, but no technical enforcement or validation of the generated summary is present.
[DYNAMIC_EXECUTION]: The skill dynamically constructs a shell command by interpolating the conversation summary directly into a command string, which could be exploited if the summary contains shell metacharacters.

claude-handoff