The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides numerous examples of CLI commands for chaos testing (e.g., aqe chaos run, aqe chaos load, aqe chaos stress). These are intended for quality engineering tasks but involve powerful system-level operations like injecting network latency, stressing CPU/memory, and terminating processes.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and report on external data sources, including system logs, metrics, and traces, as evidenced by the ChaosReport interface in SKILL.md. This creates a potential surface for indirect prompt injection if the ingested data contains malicious instructions intended to influence the agent's summary or findings.
Ingestion points: External logs, metrics, and traces are ingested to generate chaos reports (specified in SKILL.md under ChaosReport).
Boundary markers: None identified in the provided instructions for delimiting external log or metric data.
Capability inventory: The skill uses aqe CLI tools for fault injection and stress testing; scripts like scripts/validate-config.json suggest validation capabilities.
Sanitization: No explicit sanitization or filtering of external log/metric content is documented before it is processed for the final report.

qe-chaos-resilience