ollygarden-insight-remediation

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill instructs the agent to ask for, read, store, and embed API keys (examples like og_sk_..., and curl -H "Authorization: Bearer $KEY") into commands/files, which requires handling secret values verbatim and risks exfiltration.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.90). The skill explicitly fetches insights and the remediation_instructions from the external OllyGarden API (e.g., https://api.ollygarden.cloud/api/v1/services/{id}/insights) and the SKILL.md workflow requires quoting and strictly following those API-provided instructions to make code changes, so untrusted third-party content can directly drive agent actions.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 1.00). The skill explicitly fetches remediation_instructions at runtime from https://api.ollygarden.cloud (e.g., https://api.ollygarden.cloud/api/v1/services and https://api.ollygarden.cloud/api/v1/services/{id}/insights, and also references https://api.ollygarden.cloud/llms.txt) and requires the agent to quote and follow those remote instructions, so fetched content directly controls agent behavior.