MCPgrade

MCP server · trust report

io.github.polterguy/magic

Published by io.github.polterguy

version-controlinfrastructuredatabase

Only 2 of 6 checks could run, which is below the bar to certify — nothing that ran raised a red flag.

Publisher
io.github.polterguy
Repository
github.com/polterguy/magic
Install
oci:docker.io/servergardens/magic-aio:v23.1.3, oci:docker.io/servergardens/magic-aio:v23.1.4, oci:docker.io/servergardens/magic-aio:v23.1.5, oci:docker.io/servergardens/magic-aio:v23.1.6, oci:docker.io/servergardens/magic-aio:v23.1.7, oci:docker.io/servergardens/magic-aio:v23.1.8, oci:docker.io/servergardens/magic-aio:v23.1.9, oci:docker.io/servergardens/magic-aio:v23.1.10, oci:docker.io/servergardens/magic-aio:v23.2.0, oci:docker.io/servergardens/magic-aio:v23.2.1, oci:docker.io/servergardens/magic-aio:v23.2.2, oci:docker.io/servergardens/magic-aio:v23.2.3, oci:docker.io/servergardens/magic-aio:v23.2.4, oci:docker.io/servergardens/magic-aio:v23.2.5, oci:docker.io/servergardens/magic-aio:v23.2.6, oci:docker.io/servergardens/magic-aio:v23.2.7, oci:docker.io/servergardens/magic-aio:v23.2.8, oci:docker.io/servergardens/magic-aio:v23.3.0, oci:docker.io/servergardens/magic-aio:v23.3.2, oci:docker.io/servergardens/magic-aio:v23.3.3, oci:docker.io/servergardens/magic-aio:v23.3.5, oci:docker.io/servergardens/magic-aio:v23.5.0, oci:docker.io/servergardens/magic-aio:v23.5.1, oci:docker.io/servergardens/magic-aio:v23.5.2, oci:docker.io/servergardens/magic-aio:v23.5.3, oci:docker.io/servergardens/magic-aio:v23.5.4, oci:docker.io/servergardens/magic-aio:v23.5.5, oci:docker.io/servergardens/magic-aio:v23.5.6, oci:docker.io/servergardens/magic-aio:v23.5.7, oci:docker.io/servergardens/magic-aio:v23.5.8, oci:docker.io/servergardens/magic-aio:v23.5.9, oci:docker.io/servergardens/magic-aio:v23.5.10, oci:docker.io/servergardens/magic-aio:v23.5.11, oci:docker.io/servergardens/magic-aio:v23.5.12, oci:docker.io/servergardens/magic-aio:v23.5.13, oci:docker.io/servergardens/magic-aio:v23.5.14, oci:docker.io/servergardens/magic-aio:v23.5.15
Versions
37
Inspected
2 of 6 checks · 35%
Scored
2026-09-04 · rubric 1.0.0

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.

6 checks2 clean4 not inspected

  • Injection surface

    unscannable25% of grade

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

    No tool descriptions, server instructions, prompts, or resources were fetched; nothing to scan for injection.

  • Supply chain

    pass25% of grade

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

    repository linkage verified

  • Credential hygiene

    unscannable15% of grade

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

    No source files, package manifest, or tool schemas were fetched; cannot assess credential handling.

  • Permission scope

    unscannable15% of grade

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

    No source files, package.json, or pyproject were fetched; cannot infer capability scope.

  • Version behavior

    pass10% of grade

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

    tool surface, capabilities, install scripts, and publisher unchanged vs prior version 23.5.14 (patch bump); no rug-pull drift.

  • Transport config

    unscannable10% of grade

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

    local stdio server: no remote transport or auth surface, and no phone-home/telemetry signal found in source. Absence of transport findings is NOT a safety guarantee — phone-home hosts can be constructed at runtime and defeat static analysis.

Version history

Each release plotted by grade against the safe line at B. A version that sinks below the line has lost its trusted standing — the shape of a rug-pull.

VersionPublishedGradeScoreInspectedChange
v23.5.15 · current2026-09-04 insufficient—2/6 · 35%±0
v23.5.142026-09-02 insufficient—2/6 · 35%±0
v23.5.122026-08-31 insufficient—2/6 · 35%±0
v23.5.112026-08-30 insufficient—2/6 · 35%±0
v23.5.102026-08-25 insufficient—2/6 · 35%±0
v23.5.92026-08-25 insufficient—2/6 · 35%±0
v23.5.82026-08-21 insufficient—2/6 · 35%±0
v23.5.62026-08-17 insufficient—2/6 · 35%±0
v23.5.12026-08-17 insufficient—1/6 · 25%±0
v23.3.52026-08-16 insufficient—2/6 · 35%±0
v23.3.32026-08-14 insufficient—2/6 · 35%±0
v23.2.72026-08-13 insufficient—2/6 · 35%±0
v23.2.62026-08-12 insufficient—2/6 · 35%±0
v23.2.32026-08-11 insufficient—2/6 · 35%±0
v23.2.12026-08-10 insufficient—2/6 · 35%±0
v23.1.102026-08-08 insufficient—2/6 · 35%±0
v23.1.92026-08-08 insufficient—2/6 · 35%±0
v23.1.82026-08-07 insufficient—2/6 · 35%±0
v23.1.62026-08-04 insufficient—2/6 · 35%±0
v23.1.42026-08-04 insufficient—2/6 · 35%±0
v23.1.32026-08-01 insufficient—1/6 · 25%—