asrai-mcp-apikey
Asrai crypto analysis MCP server — API key auth for ADVANCED plan users.
This report shows the evidence for asrai-mcp-apikey@1.10.0. 0 of 20 known versions are blocked or flagged. See all versions.
AI Security Review
scanned 21h ago · by lpm-firewall-aiThe source behavior matches the package’s stated purpose: it forwards user-requested crypto-analysis calls and their Asrai key to the Asrai service. The agent-skill writes occur only through the explicit installer command and are limited to the package’s own skill files.
Decision evidence
public snapshot- The manifest exposes `install-skill` as a user-run command and declares no install lifecycle hook.
- The skill installer copies the package’s skill files into detected agent skill directories when explicitly run.
- The stdio bridge resolves an `ASRAI_KEY` or `--key` value and connects to the package’s documented Asrai MCP service.
- The `positions` tool forwards exchange credentials to the Asrai API for the positions feature.
- The bridge sends its key to `mcp-api.asrai.me`, the service this package is designed to access.
- The installer writes only the package skill and endpoint reference files under the `asrai` skill directory.
- No source reviewed showed automatic collection or transmission of unrelated credentials or host data.
Source & flagged code
5 flaggedManifest-trigger-reachable source writes behavior-bearing configuration into a user or project AI-agent control surface.
bin/install-skill.jsView on unpkgThe skill installer copies the package’s skill files into detected agent skill directories when explicitly run.
bin/install-skill.jsView on unpkg · L48Findings
Affected versions and remediation
This report applies to asrai-mcp-apikey@1.10.0.
The versions list below shows the verdict for each recorded version.
- Review the evidence and your use of asrai-mcp-apikey@1.10.0 before allowing it. Restrict the permissions described in this report.
- Choose an independently verified alternative or release. This report does not establish that other versions are safe.
Evidence last updated: .
Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.