Gridzen Verification

Gridzen Developer Kit 0.7.0

The hosted developer interface supports Vietnamese alongside English, Spanish and Simplified Chinese: https://gridzen.ai/developers/?lang=vi . UI language does not enable live checks or country coverage. See UI language maintenance.

Choose direct integration or GridZen for your task · Production and cost checklist

Single-task entry points: tool catalog · setup and boundaries.

python3 -m venv .venv

.venv/bin/python -m pip install gridzen-developer-kit .venv/bin/gridzen tools .venv/bin/gridzen setup --client codex --product clabe-format --project .

Review the preview, then add --apply. No key is needed for local tools.

Start immediately without signup

Open GridZen instant trial to run synthetic signup-phone and identity-onboarding responses without an account, API key or approval. Success, rejection and unavailable scenarios make no upstream calls and verify no real person. Agents can use the separate key-free trial MCP, installed with mcp==2.3.0 and httpx==0.28.1 in a virtual environment.

For persistent test records and supplier sandbox callbacks, connect a verified merchant sandbox account. Production access is capability-specific: eligible routes may offer bounded real trials, while others remain unavailable. Read the authenticated capability conditions; published prices do not enable paid purchases. This documentation update does not change PyPI or the official MCP Registry version.

Fourteen Skills: the original six research/workflow Skills plus eight single-task profiles. Three new tools run offline; five merchant profiles call the existing authenticated GridZen API only with separate authorization, current availability and funding. The anonymous remote MCP retains five research/synthetic tools and never performs live verification. Research covers 198 countries/territories as a dated 2026-10-07 snapshot, not live service coverage.

中文指南 · Guía en español

Runnable task guides and distribution

Chinese and Spanish article files · Demo video · Smithery listing.

Public examples are synthetic. Approved prices and per-product availability are published at https://gridzen.ai/pricing/; live payments remain disabled. Request an approved pilot, without sending identity documents or credentials to public MCP tools.

Per-check plan and activation requirements · JSON

Try it

Public playground: https://gridzen.ai/developers/

Download gridzen-developer-kit.zip, extract it, then:

python3 -m venv .venv

. .venv/bin/activate python -m pip install './gridzen-developer-kit[mcp]' gridzen coverage --country ID --capability bank_account_match gridzen plan --country ID --event payout gridzen simulate --country ID --capability bank_account_match --scenario timeout python gridzen-developer-kit/examples/payout.py

Python 3.11+. CLI and MCP use the bundled offline catalog and fixtures; the Python HTTP client/example calls the public sandbox. Dependencies are downloaded only during installation. API reference: https://gridzen.ai/developers/api/openapi.json .

MCP

Register a local stdio server in your MCP client. Use the absolute path to the installed executable; configuration syntax varies by client. A conventional mcpServers configuration is:

{"mcpServers":{"gridzen":{"command":"/absolute/path/to/.venv/bin/gridzen-mcp","args":[]}}}

Tools: get_coverage, plan_verification, create_sandbox_verification, get_sandbox_verification, explain_result. All operate on local research or synthetic fixtures, with read-only/idempotent annotations. Public research/sandbox MCP endpoint: https://gridzen.ai/developers/mcp (Streamable HTTP, no account or API key). It has the same five tools and zero live providers. Remote usage sends the documented arguments to Gridzen; local stdio is offline by default. Limits: 120 requests/minute per client address and 600/minute overall, returning HTTP 429 with Retry-After.

Skills

Copy the individual directories from skills/ into your compatible agent's project or user skill directory. Each entrypoint is SKILL.md; preserve its references/ directory. No marketplace listing or automatic registration is implied.

HTTP SDK

from gridzen_developer.client import Gridzen

client = Gridzen() plan = client.plan('ID', 'payout') result = client.simulate('ID', 'bank_account_match', 'not_found') assert result['status'] == 'inconclusive' assert result['verified'] is False

Only documented country/event/capability/scenario fields are accepted. Never send identity numbers, bank accounts, documents or credentials to the sandbox. Simulation IDs encode fixtures and are not unique transaction identifiers. A synthetic match cannot authorize real users or payments. Research evidence is distinct from Gridzen availability or commercial permission.

Repository checkout: local service and checks

python -m pip install '.[test,server]'

python -m pytest -q python scripts/check_mcp.py uvicorn gridzen_developer.server:app --host 127.0.0.1 --port 8030

Production uses an independent container at localhost:8030 proxied only under /developers/. See deploy/compose.yml and deploy/developers.conf. Removing that proxy file and stopping this container rolls back the preview; the existing API and Agent Company stacks are separate.

Repository checkout: build distributions

python scripts/build_downloads.py creates allowlisted developer-kit and Skills ZIPs; it excludes tests, deployment files, private credentials and itself from recursive download packaging. python scripts/build_catalog.py /docs/research/identity-verification/2026-10-07 refreshes the public research projection. Main repository report files remain the provenance source.

Public distribution

Source: https://github.com/immurray/gridzen-developer-kit . Install the Skills:

npx skills add immurray/gridzen-developer-kit

Install the tagged Python/MCP release from GitHub (Python 3.11+ and Git):

python -m pip install "gridzen-developer-kit[mcp] @ git+https://github.com/immurray/gridzen-developer-kit.git@v0.6.0"

PyPI 0.6.0 is published and verified. Install with python -m pip install gridzen-developer-kit. See the release tracker for publication and installation evidence.

Remote client configuration (conventional JSON; client syntax may vary):

{"mcpServers":{"gridzen":{"type":"http","url":"https://gridzen.ai/developers/mcp"}}}

For Claude Code: claude mcp add --transport http gridzen https://gridzen.ai/developers/mcp. For Cursor: add the remote URL in its MCP settings. For portable Agent Plugins, use the packaged plugin.json and mcp.json; the portable transport identifier is streamable-http. Marketplace admission is separate from compatibility.

Original code and Skills: MIT. Research provenance and third-party rights: NOTICE.md. Support: open@gridzen.ai .

Marketplace availability

Source and previous release downloads are public. Existing Registry and Skills.sh listings cover the research service; new product profiles are not separate approved listings. Docker catalog submission is awaiting review. See the publication ledger for exact status and remaining platform steps.

14 bundled Skills (0.7.0)

python -m pip install gridzen-developer-kit installs the CLI, stdio MCP dependencies and all 14 Skill directories.

The 14 Skills comprise 6 research/workflow Skills and 8 single-task profiles. Of the 8 profiles, 3 run offline and 5 merchant profiles require separate authorization. Five public MCP tools provide research/synthetic operations. No live identity verification. No extra MCP dependency installation is required. Version 0.6.0 is published on PyPI.

gridzen coverage --country MX

gridzen skills gridzen-mcp

Connect gridzen-mcp as your client's stdio server, or use the remote endpoint https://gridzen.ai/developers/mcp. gridzen skills prints the installed directories; copy a complete directory into your assistant's configured Skills directory to activate it. Installing a wheel does not automatically activate a client's Skills.

The existing select/integrate/explain Skills remain available. Added offline workflows, version 1.0.0: mexico-pilot-scoper, payout-policy-designer, and provider-rights-readiness. Each includes MIT, README, complete response fixtures and agent acceptance prompts. They make no third-party calls and grant no real verification, legal approval or regulatory conclusion.

Configure your assistant (v0.6.0)

python3 -m venv .venv

. .venv/bin/activate python -m pip install --upgrade gridzen-developer-kit gridzen setup --client all --project . gridzen setup --client all --project . --apply

Use --client for Codex, Claude Code/Desktop, Cursor, VS Code/Copilot, Copilot CLI, Gemini CLI, Cline, Roo Code or OpenCode. Setup previews by default, merges existing unrelated settings, backs up changed configurations and refuses differing Gridzen entries or Skill files. No global client settings are modified. Native clients require workspace trust; Desktop and Cline require manual MCP import. Desktop produces separately uploadable Skill ZIPs. See client-specific instructions and actual compatibility evidence. Configuration preparation is not model acceptance or real verification.

Hosted MCP aggregate attribution

Hosted calls count successful tool invocations, not customers. Existing totals remain unchanged. The hosted service records UTC day, tool name, one source category and a count; it never persists request arguments or client addresses. Historical and unlabelled calls remain unknown. Local stdio is offline by default.

Explicit acceptance requests to the loopback-published service may set X-Gridzen-Purpose: internal_test. Only configured local peers with no forwarded client headers are classified as internal_test. The public reverse proxy must overwrite X-Real-IP; publicly supplied labels are only self_reported_test. An unlabelled public request is not an authenticated external customer. The GRIDZEN_MCP_LOCAL_TEST_PEERS deployment setting must match the container gateway and must never contain public client addresses. Source counts use the existing 90-day retention and fail-open behavior. This hosted change does not publish a new PyPI version or enable any live provider.

Customer tasks and voluntary feedba