monid

Fail

Audited by Snyk on Nov 15, 2026

Risk Level: CRITICAL
Full Analysis

CRITICAL E004: Prompt injection detected in skill instructions.

  • Potential prompt injection detected (medium risk: 0.30). The skill is an openly documented Monid CLI guide with no obfuscated code, covert triggers, impersonated authority, or pretexts, but it has medium-risk behavior: it has the agent auto-replace its own skill file from a remote URL and send feedback and dispute reports containing user request context to the vendor at the end of tasks.

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

  • Third-party content exposure detected (low risk: 0.10). The required workflow runs third-party scraping endpoints (e.g. Twitter/LinkedIn/Google Maps via Apify) on agent-chosen queries and then reads their outputs, so outsider-authored posts are ingested only as search results the agent actively requests, not from a feed an outsider can push into.

MEDIUM W009: Direct money access capability detected (payment gateways, crypto, banking).

  • Direct money access detected (medium risk: 0.30). Operation: monid run executes paid endpoints that debit the user's prepaid Monid wallet ("Monid runs spend the user's Monid balance"), and monid topup and wallet-balance blocks handle funding. Running a paid endpoint is a purchase, so it is a direct operation. The "service costs" exclusion covers an agent paying for its own service call, but this skill debits the wallet to buy data from third-party endpoints on the user's behalf, so the exclusion does not apply. Control: each run requires a Monid API key that the service validates (no key gives "No active API key", and an invalid key gives 401 "API key is expired or invalid"). This is service-enforced credential validation for the payment route, so it qualifies as an established control. The skill's own instructions to use the user's tools first are agent-followed and are not counted. Because a qualifying control exists, the direct route is medium, not high.

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

  • Attempt to modify system services in skill instructions detected (low risk: 0.35). The skill mainly asks for a global npm install of a CLI and API key setup, which are acceptable, but it also tells the agent to overwrite its own skill file with the latest copy from a remote URL and keep it enabled, a persistent change to the agent's configuration.

Issues (4)

E004
CRITICAL

Prompt injection detected in skill instructions.

W011
MEDIUM

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

W009
MEDIUM

Direct money access capability detected (payment gateways, crypto, banking).

W013
MEDIUM

Attempt to modify system services in skill instructions.

Audit Metadata
Risk Level
CRITICAL
Analyzed
Nov 15, 2026, 06:15 AM
Issues
4