The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute several shell commands, including kongctl, awk, grep, and env. These are used for intended diagnostic purposes:
kongctl version and kongctl help for environment verification.
env | grep '^KONGCTL_' to inspect configuration overrides (references/auth-and-scope-checks.md).
awk scripts to parse help output for resource discovery (references/resource-discovery.md).
[INDIRECT_PROMPT_INJECTION]: The skill has an indirect injection surface because it processes output from the kongctl CLI, which may contain data from an external environment (Konnect).
Ingestion points: Data enters the agent's context through the output of kongctl get commands in JSON, YAML, or text formats.
Boundary markers: None explicitly defined for the external tool output.
Capability inventory: The skill uses kongctl, awk, grep, and env (SKILL.md, references/resource-discovery.md).
Sanitization: No specific sanitization or escaping is mentioned, though the skill emphasizes summarizing results to reduce data volume (references/output-shaping.md).
[CREDENTIALS_SAFE]: The skill explicitly includes security instructions to handle credentials safely. It states, "Never echo, log, or commit token values" and "Never echo or paste token values back into logs or responses," mitigating the risk of accidental exposure during authentication checks (SKILL.md, references/auth-and-scope-checks.md).

kongctl-query