slo-sli

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt embeds a plaintext credential (OPENSEARCH_PASSWORD = "My_password_123!@#") directly in the skill instructions, which exposes a secret in the model context and risks it being echoed or exfiltrated.

HIGH W008: Secret detected in skill content (API keys, tokens, passwords).

• Secret detected (high risk: 1.00). I inspected the skill content for literal credentials. The table under "Connection Defaults" contains:
• OPENSEARCH_PASSWORD: My_password_123!@#

This is a concrete, non-placeholder password (mixed case, digits, special characters) and therefore meets the definition of a secret (a literal credential that could grant access to OpenSearch). It should be treated as a sensitive credential if used in real deployments.

Items ignored:

• OPENSEARCH_USER: "admin" — common default/low-entropy username, treated as non-secret.
• AWS variables (AWS_ACCESS_KEY_ID, AWS_SECRET_ACCESS_KEY) appear only as environment variable names/placeholders; no actual keys are present, so they are ignored per the documentation-placeholder rule.
• Other values like localhost endpoints and curl commands contain no other literal high-entropy credentials.

Therefore I flag the OpenSearch password as a secret.