io.github.LE-VAI/designesy-org

What we checked

Six static checks, weighted by risk. Every result reflects only what could be observed in the published package and repository — never intent.

• Injection surface

  warn25% of grade

  Tool descriptions/manifest scanned for instruction-injection patterns (imperatives at the model, hidden text, 'ignore previous', data-exfil URLs).

  tool "designesy_score" param "url" description: off-vendor URL in a description (www.designesy.org); capability text should not name where data goes. tool "designesy_drift_score" param "url" description: off-vendor URL in a description (www.designesy.org); capability text should not name where data goes. tool "designesy_readiness_score" param "url" description: off-vendor URL in a description (www.designesy.org); capability text should not name where data goes. tool "designesy_guardrails" param "url" description: off-vendor URL in a description (www.designesy.org); capability text should not name where data goes. tool "designesy_monitor_score" param "url" description: off-vendor URL in a description (www.designesy.org); capability text should not name where data goes.

• Supply chain

  pass25% of grade

  Package provenance: namespace verification, repo linkage, maintainer count, account age, postinstall scripts, typosquat distance.

  repository linkage verified

• Credential hygiene

  pass15% of grade

  How the server takes secrets (env vs plaintext config vs hardcoded); secrets appearing in tool schemas.

  No hardcoded secrets, secret-shaped tool parameters, or leaked keys found. PASS = no static red flag; static analysis cannot prove the code honors env-based secret handling at runtime.

• Permission scope

  warn15% of grade

  Declared tools vs. breadth (filesystem, network, exec); flags shell-exec and unbounded filesystem access.

  has both shell/exec and network egress — capable of download-and-run, though no such path was observed (designesy_mcp_server.py: subprocess.* call; designesy_mcp_server.py: outbound HTTP/socket call). discloses shell/exec capability: designesy_mcp_server.py: subprocess.* call Capability DISCLOSURE, not a verdict: static analysis sees the primitive is present and reachable, not whether its use is attacker-controlled.

• Version behavior

  fail10% of grade

  Diff of tool definitions between versions; new permissions or changed descriptions in a patch release (the postmark-mcp class).

  vs prior 1.11.0 (minor bump): tool "designesy_catalog" description changed (wording only, no injection pattern); tool "designesy_score" description changed (wording only, no injection pattern); tool "designesy_tokens_score" description changed (wording only, no injection pattern); tool "designesy_a11y_score" description changed (wording only, no injection pattern); tool "designesy_drift_score" description changed (wording only, no injection pattern); tool "designesy_report" description changed (wording only, no injection pattern); new capability primitive(s) in source: subprocess; new hardcoded outbound host/BCC destination(s): 127.0.0.1, github.com, registry.modelcontextprotocol.io, www.designesy.org.

• Transport config

  fail10% of grade

  Remote servers: TLS and auth mode (none/token/OAuth). Local servers: whether the manifest indicates it phones home.

  remote MCP endpoint declares NO authentication (unauthenticated remote — nginx-ui/CVE-2026-33032 class)