The Agent Skills Directory

[DATA_EXFILTRATION]: Accesses sensitive system information containing credentials.
The skill instructs the agent to read /proc/1/environ using the command cat /proc/1/environ | tr '\0' '\n'. This file contains the environment variables of the process with PID 1, which commonly includes sensitive secrets, API keys, and database connection strings.
[INDIRECT_PROMPT_INJECTION]: Vulnerability surface via untrusted data ingestion.
Ingestion points: The skill reads application logs to check for startup errors and parses file contents using grep to verify revisions.
Boundary markers: None are specified to prevent the agent from following instructions embedded in log messages or file contents.
Capability inventory: The skill utilizes Bash, Read, and Grep tools, which could be leveraged if an injection is successful.
Sanitization: There is no mention of sanitizing or escaping the log data or file contents before the agent processes them.

deploy-landed