magpie-trustoken-auth@0.3.7
Trustoken / AiPy for magpie — phone-code sign-in, the account's live model list, 灵石 balance, auto check-in and daily-task rewards. DeepSeek / Qwen / GLM / Doubao / Kimi / Hunyuan.
AI Security Review
scanned 5d ago · by lpm-firewall-aiThe package’s default automatic check-in path can spend the user’s model quota by creating AiPy tasks or sending fallback model requests with the user’s keys. Because this is automatic resource spending, the package meets the malicious verdict rule despite the behavior being tied to its stated daily-rewards feature.
Decision evidence
public snapshot- The package sets `magpie.autoCheckin` to true, enabling automatic check-in by default.
- When check-in finds unfinished daily tasks, it creates AiPy tasks using the cheapest model; if that path is unavailable, it sends chat requests with the account API key.
- The task-creation helper reads AiPy’s local API key and sends it to AiPy’s local task endpoint. Creating those model tasks can consume the user’s quota.
- The package describes itself as a Trustoken/AiPy sign-in and daily-rewards plugin; check-in and task completion serve that purpose.
- The entry point assembles hooks without performing account actions at import time.
- The OAuth flow sends the session cookie to the Trustoken service during the user-started login flow; no unrelated credential destination was identified.
- The manifest has no install or lifecycle scripts.
Source & flagged code
11 flaggedSource file is highly similar to a previously finalized malicious package; route for source-aware review.
scripts/score.mjsView on unpkgFindings
Affected versions and remediation
This report applies to magpie-trustoken-auth@0.3.7.
See all versions of magpie-trustoken-auth for other recorded verdicts.
- Avoid installing magpie-trustoken-auth@0.3.7. Remove it from direct dependencies and check your lockfile for transitive copies.
- Choose an independently verified alternative or release. This report does not establish that other versions are safe.
- If this version ran, investigate the affected machine and build environment. Rotate credentials it could access and rebuild from a trusted environment.
Evidence last updated: .
Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.