arete-build
Warn
Audited by Snyk on May 10, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.80). This skill explicitly instructs fetching external, public data (e.g., SKILL.md "Enriching with off-chain data" which uses the TokenMetadata resolver/DAS API and the DSL reference "URL Resolver" that fetches JSON from arbitrary HTTP endpoints or metadata URIs), and those untrusted third‑party payloads are ingested and used to compute/transform entity fields (e.g., decimals, ui_amount, configuration), so external content can materially affect agent decisions and behavior.
MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).
- Potentially malicious external URL detected (high risk: 0.90). The skill includes runtime installation commands that fetch and execute remote installers (curl ... | sh https://sh.rustup.rs and downloading/executing https://win.rustup.rs/x86_64), which are external URLs used at runtime that run remote code and are required to provision the Rust toolchain.
Issues (2)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
W012
MEDIUMUnverifiable external dependency detected (runtime URL that controls agent).
Audit Metadata