MCP server · trust report

sh.handover/handover

2 checks that ran failed — credential hygiene and version behavior — so this release can’t be trusted as-is.

Publisher
sh.handover
Repository
github.com/44-pixels/handover-mcp
Install
remote only
Versions
3
MCP revision
2025-11-25 · superseded
Inspected
5 of 6 checks · 85%
Scored
2026-09-03 · 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 checks3 clean2 flagged1 not inspected
  • Injection surface

    pass25% of grade

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

    Scanned 489 text field(s); no instruction-injection patterns found. PASS = no static red flag. A server can serve different tool descriptions at runtime (the rug-pull class), which static scanning cannot see.

  • Supply chain

    pass25% of grade

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

    repository linkage verified

  • Credential hygiene

    fail15% of grade

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

    tool "handover.source.create" has a secret-shaped input parameter "secretEnv" — secrets must never be tool parameters (the model would supply/handle the credential). tool "handover.destination.create" has a secret-shaped input parameter "secretEnv" — secrets must never be tool parameters (the model would supply/handle the credential).

  • 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

    fail10% of grade

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

    vs prior 1.16.0 (minor bump): tool "list_contexts" description changed (wording only, no injection pattern); tool "get_context" description changed (wording only, no injection pattern); tool "pull_context" description changed (wording only, no injection pattern); tool "deploy_context" description changed (wording only, no injection pattern); tool "update_context" description changed (wording only, no injection pattern); tool "manage_folders" description changed (wording only, no injection pattern); tool "export_okf" description changed (wording only, no injection pattern); tool "import_okf" description changed (wording only, no injection pattern); tool "handover.spaces" gained input property(ies): name, description, kind, access, status; tool "handover.resources" gained secret/exfil-shaped input(s): note; tool "handover.requests" gained input property(ies): title, description, status, priority, assigneePrincipalId, sourceHandoverId; tool "get_context_index" description changed (wording only, no injection pattern); tool "get_context_health" description changed (wording only, no injection pattern); tool "get_context_history" description changed (wording only, no injection pattern); tool "list_products" description changed (wording only, no injection pattern); tool "get_product" description changed (wording only, no injection pattern); tool "read_product_file" description changed (wording only, no injection pattern); tool "get_product_health" description changed (wording only, no injection pattern); tool "get_product_index" description changed (wording only, no injection pattern); tool "get_product_history" description changed (wording only, no injection pattern); tool "find_assets" description changed (wording only, no injection pattern); tool "find_assets" gained input property(ies): query, limit; tool "list_backlog" description changed (wording only, no injection pattern); tool "delete_context" description changed (wording only, no injection pattern); tool "create_product" description changed (wording only, no injection pattern); tool "create_product" gained input property(ies): name, description; tool "delete_product" description changed (wording only, no injection pattern); tool "update_product_knowledge" description changed (wording only, no injection pattern); tool "update_product_knowledge" gained input property(ies): content_type, encoding, updated_by; tool "upload_product_files" description changed (wording only, no injection pattern); tool "delete_product_file" description changed (wording only, no injection pattern); tool "add_to_backlog" description changed (wording only, no injection pattern); tool "add_to_backlog" gained input property(ies): priority; tool "process_backlog" description changed (wording only, no injection pattern); tool "process_backlog" gained input property(ies): limit; tool "delete_backlog_item" description changed (wording only, no injection pattern).

  • Transport config

    pass10% of grade

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

    remote endpoint "https://handover.sh/api/mcp" uses TLS (https); remote endpoint "https://handover.sh/api/mcp?profile=core" uses TLS (https); remote endpoint uses OAuth (strongest auth mode)

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
v1.18.0 · current2026-08-04 B755/6 · 85%±0
v1.16.02026-08-03 B854/6 · 75%