substore-openclash

Fail

Audited by Snyk on Aug 18, 2026

Risk Level: HIGH
Full Analysis

HIGH W007: Insecure credential handling detected in skill instructions.

  • Insecure credential handling detected (high risk: 1.00). The prompt instructs the operator to read secret tokens (SubStore/AIO token, OpenClash secret, device MIIO token) and to embed them verbatim into curl/node commands (e.g., Authorization header and TOKEN variable), which requires the LLM to handle/output secret values directly.

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

  • Third-party content exposure detected (medium risk: 0.30). OpenClash on the gateway fetches SubStore’s generated AIO Clash YAML from http://<nas-ip>:3001/sub-api/api/file/AIO (pipeline output), so any outsider-authored text contained in upstream SubStore subscriptions/collections can flow into runtime YAML consumed by the gateway/Clash without selecting a specific item first.

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


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

  • Secret detected (high risk: 1.00). I flagged a high-entropy, literal device token present in the documentation. The 32-hex string (000662ac87cd9dc57bc36b5a9816f917) is used as the MiIO device token in the local-miio verification and is a real secret (grants local device access). I ignored documentation placeholders like <SUBSTORE_AIO_TOKEN> and other non-secret examples.

MEDIUM W013: Attempt to modify system services in skill instructions.

  • Attempt to modify system services in skill instructions detected (high risk: 1.00). The skill includes explicit, repeatable instructions to SSH into hosts and modify persistent system and container files (editing sub-store.json, patching/homebridge node_modules, copying files into containers and restarting services), which changes the machine state and requires privileged access.

Issues (5)

W007
HIGH

Insecure credential handling detected in skill instructions.

W011
MEDIUM

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

W012
MEDIUM

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

W008
HIGH

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

W013
MEDIUM

Attempt to modify system services in skill instructions.

Audit Metadata
Risk Level
HIGH
Analyzed
Aug 18, 2026, 03:51 PM
Issues
5
Security Audit — snyk — substore-openclash