The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to use a local CLI tool named ochist to perform its primary functions, including listing sessions, searching content via grep, and retrieving message parts. This involves executing shell commands with various flags and arguments.
[PROMPT_INJECTION]: The skill is a surface for indirect prompt injection because it ingests conversation history from previous sessions. If those past sessions contained untrusted content or malicious instructions that were logged, re-reading that content via ochist part or ochist grep could cause the agent to follow instructions embedded within the historical data.
Ingestion points: Data enters the agent's context through commands like ochist part <part_id> and ochist show <session> --full.
Boundary markers: No explicit boundary markers or XML tags are suggested to wrap the historical content to prevent it from being interpreted as instructions.
Capability inventory: The agent can use the output of history retrieval to inform decisions and potentially trigger other tools available in its environment.
Sanitization: The skill does not describe any sanitization or filtering of the historical text before it is processed by the agent.

agent-history