Verdict QA MCP Server
Your test suite is green. Verdict found a defect that had lived 4,595 days. Verdict is a QA agent that does not fix, does not flatter, and does not forget. It measures before it judges - the harness runs your gates, hashes every line a finding cites, re-runs the guarding test at the old commit and the new one - and it keeps a memory: every run is a delta against the last, findings age, regressions rank first, and…
People who work with claude code plugin, flaky tests and release gate and want it reachable from Claude, Cursor, VS Code, or another MCP client. The project is written in Python.
VERIFIED ACTIVE
LAST COMMIT 2026-09-28 · ★ 2 · #35 OF 321 MAINTAINED MEMORY · VERIFIED 2026-09-30
MIT · Python servers · how we verify → /methodology
01 · Install Verdict QA
before you install - you'll need
The README does not document required environment variables for a basic install.
Claude Code
claude mcp add artjack-verdict -- uvx verdict-qa-mcp Claude Desktop / Cursor / VS Code - add to config
{
"mcpServers": {
"artjack-verdict": {
"command": "uvx",
"args": [
"verdict-qa-mcp"
]
}
}
} Same JSON for Cursor. For VS Code, rename the top-level key from `mcpServers` to `servers`.
Using another client? Same JSON, different key
Claude Desktop · mcpServers
Cursor · mcpServers
VS Code · servers
Windsurf · mcpServers
Zed · context_servers
Cline · mcpServers
Roo Code · mcpServers
Continue · mcpServers
LibreChat · mcpServers
Gemini CLI · mcpServers
Codex CLI · mcp_servers
Full setup guides: every client.
02 · Evidence
Security posture
What to check before giving this server access to your agent - from the registry, GitHub, and our own probes. We don't score safety; we show what's verifiable.
runs as local process (stdio) - runs on your machine with your user's permissions
repo age created 2026-08-25 - young repo, little track record yet
license MIT - declared in the repository
pypi package verdict-qa-mcp - check the name against the project README before installing (PyPI has no namespace ownership)
registry namespace io.github.ArtJack is GitHub-verified and matches the repo owner
03 · What Verdict QA can do
Prose above is summarized from the project's README and registry record - no invented capabilities.
Latest releases
v0.90.2 · 2026-09-23
A refused finalize kept the questions it asked. verdict-finalize folded the judgment's · questions into questions.json - an id minted for each, every waiting answer marked · acknowledged - before the validator had…
v0.90.1 · 2026-09-19
0.90.0's runner fix leaked into the project's tests. To let a runner started from a checkout · find its own harness, 0.90.0 put the runner's source root on PYTHONPATH for the harness child · - and every process that…
v0.90.0 · 2026-09-18
Every run nobody asked for spends zero Claude tokens. verdict-run --on-drift · {model,local,none} decides what a night does when the sweep is blocked: spend a Claude run · (the default, unchanged), hand it to…
04 · Who maintains Verdict QA
Verdict QA is maintained by artjack. It's the only MCP server we track from this author; the repo dates to Aug 2026.
05 · Facts
- repository
- github.com/artjack/verdict
- category
- memory - ranked #35 of 321 actively-maintained memory servers as of 2026-09-30.
- release cadence
- 10+ releases in the last 90 days (latest 2026-09-23)
- registry
- io.github.ArtJack/verdict (active, first published 2026-09-01 · 18 versions)
- packages
- pypi:verdict-qa-mcp
06 · Verdict QA FAQ
What is Verdict QA?
Your test suite is green. Verdict found a defect that had lived 4,595 days. Verdict is a QA agent that does not fix, does not flatter, and does not forget. It measures before it judges - the harness runs your gates, hashes every line a finding cites, re-runs the guarding test at the old commit and the new one - and it keeps a memory: every run is a delta against the last, findings age, regressions rank first, and…
Is Verdict QA still maintained?
Yes - as of 2026-09-30, its last commit was 2026-09-28 and it shipped 10+ releases in the last 90 days. We re-verify nightly.
How do I install Verdict QA?
Run `uvx verdict-qa-mcp`. The README does not document required environment variables for a basic install. You can also paste the ready-made client config above.
Does Verdict QA run locally?
Yes - it's a stdio server: it runs on your machine (via uvx) with your user's permissions. Your data stays local unless the server itself calls external APIs.
07 · Alternatives to Verdict QA
Alternatives to Verdict QA
Maintained memory servers if Verdict QA isn't the fit.
- Codebase Memory Codebase knowledge graph for AI agents - 162 languages, sub-ms queries, 99% fewer tokens. ★ 45,519 · 2026-09-28
- Superlog Open-source agent that observes and fixes your application. Query logs, traces, metrics, incidents. ★ 1,457 · 2026-09-24
- Tradememory Protocol Tamper-evident decision audit trail and outcome-weighted memory for AI trading agents. ★ 1,425 · 2026-09-14
- Persome Local-first personal memory and model server for trusted MCP agents on macOS. ★ 1,324 · 2026-09-25
- Neo4j Memory MCP Neo4j Knowledge Graph Memory Server ★ 986 · 2026-09-09
- Projectmem Coding agent memory - one local MCP server for every project. Warns before repeating failed fixes. ★ 846 · 2026-09-30
Pairs well with
Servers that cover what Verdict QA doesn't - only shown when the pairing reason fits the companion.
More memory MCP servers · Bhived MCP · Leteo · Genesys Memory (astrix-labs) · Archivist AI · Logic Server
More Python MCP servers · Hefesto AI · Schema Brain · 2Captcha MCP · Statistics Canada MCP Server · Ascript MCP · see all