The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses local shell commands to interact with the system, specifically invoking the openlogs and ol CLI tools. It provides instructions to run commands such as openlogs tail, openlogs bun dev, and ol npm run dev to retrieve logs and manage development processes.
[INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and interprets server logs, which are often influenced by external, untrusted inputs (e.g., HTTP request data logged by a server).
Ingestion points: Data enters the agent's context through the output of openlogs tail and by reading files in the .openlogs/ directory, such as latest.txt.
Boundary markers: There are no boundary markers or instructions to the agent to treat the log content as data rather than instructions, increasing the risk that the agent might obey commands embedded in the logs.
Capability inventory: The agent has the capability to execute shell commands and read files based on the skill's instructions.
Sanitization: No sanitization or filtering of the log content is performed before it is analyzed by the agent.
[DATA_EXFILTRATION]: The skill accesses local server logs, which frequently contain sensitive information such as session tokens, internal file paths, environment variables, or personally identifiable information (PII) if the application's logging is not properly configured.

openlogs-server-logs