Identifies specific failure modes where verification tools incorrectly report success by searching a registry of known blind spots.
Also known as simulacra/verifyfirst, verifyfirst-mcp
- Stars
- 0
- Tools
- —
- Downloads, 30 days
- 174
- Latest version
- 1.8.0
- Last commit
- Aug 24, 2026
- License
- —
What is verifyfirst MCP?
verifyfirst identifies specific failure modes where verification tools incorrectly report success by searching a registry of known blind spots.
Summary generated from the project's README
About verifyfirst MCP server
Identifying specific failure modes where verification tools incorrectly report success is the primary function of this server. It helps developers find bugs that hide behind successful exit codes, 200 status responses, or seemingly correct screenshots. By accessing a registry of 48 known blind spots, users can discover why an instrument might be providing false confidence during testing or deployment.
Developers can use this tool locally via uvx or pip, integrating it with Claude Code or any MCP client. The server operates using a bundled registry of observed and documented failures, though it can connect to a live version using the --remote flag. It requires no external dependencies beyond the Python standard library. The registry is organized by the instrument used, such as playwright, curl, or systemctl, rather than by the underlying technology.
Specific tools include from_symptom to investigate vague issues, blind_spots to find what a specific instrument misses, and list_instruments to see the six available categories. Users can also perform substring searches or retrieve specific entries like NS-001 to understand the exact checks required to avoid these pitfalls.
What you can do with it
- 1Investigate vague deployment symptomsDescribe a situation where a process appears to succeed but produces incorrect results to find matching failure modes.
- 2Audit tool reliabilityCheck which specific errors a particular testing instrument is unable to detect.
- 3Verify work completionAccess a structured checklist to ensure a task is truly finished and not just reporting success.
Written from the project's README. Every use case is checked against a quote from it.
How to install verifyfirst MCP
Formats verified 10 Nov 2026Add to CursorAdd to GooseInstall verifyfirst MCP in Claude Code
$ claude mcp add verifyfirst -- uvx verifyfirst-mcpInstall verifyfirst MCP in Cursor
~/.cursor/mcp.json
{
"mcpServers": {
"verifyfirst": {
"command": "uvx",
"args": [
"verifyfirst-mcp"
]
}
}
}Install verifyfirst MCP in VS Code
.vscode/mcp.json
{
"servers": {
"verifyfirst": {
"type": "stdio",
"command": "uvx",
"args": [
"verifyfirst-mcp"
]
}
}
}Install verifyfirst MCP in VS Code CLI
$ code --add-mcp '{"name":"verifyfirst","type":"stdio","command":"uvx","args":["verifyfirst-mcp"]}'Install verifyfirst MCP in Claude Desktop
claude_desktop_config.json
{
"mcpServers": {
"verifyfirst": {
"command": "uvx",
"args": [
"verifyfirst-mcp"
]
}
}
}Install verifyfirst MCP in Codex
$ codex mcp add verifyfirst -- uvx verifyfirst-mcpInstall verifyfirst MCP in Codex config.toml
~/.codex/config.toml
[mcp_servers.verifyfirst] command = "uvx" args = ["verifyfirst-mcp"]
Install verifyfirst MCP in Gemini CLI
$ gemini mcp add -s user verifyfirst uvx -- verifyfirst-mcpInstall verifyfirst MCP in Gemini settings.json
~/.gemini/settings.json
{
"mcpServers": {
"verifyfirst": {
"command": "uvx",
"args": [
"verifyfirst-mcp"
]
}
}
}Automated checks
Ran Nov 15, 2026Full report →Passed our automated checks. That is not a security audit: read the code and give the server the narrowest access it needs. How checks work · What to fix for more points
- Source repository linked and reachablesimulacra/verifyfirst is public and reachablegithub.com/simulacra/verifyfirst ↗Pass
- Package names this repository as its sourceverifyfirst-mcp names simulacra/verifyfirst as sourcepypi.org/project/verifyfirst-mcp/ ↗Pass
- Build provenance (Sigstore attestation)No build attestation published for this releaseNot checked
- Open source license (OSI)No licenseFinding
- Commit or release in the last yearLast commit or release 2026-08-24github.com/simulacra/verifyfirst ↗Pass
- OpenSSF ScorecardNot checked
- Package name not close to a popular packageNo popular package one edit away from verifyfirst-mcpPass
- No critical advisory in the package or its dependencies1 resolved dependencies checked; no critical advisoryosv.dev ↗Pass
- No hidden instructions to the model0 tool descriptions and the README read; no override phrases, credential paths or hidden charactersPass
- Destructive tools are labelledNot checked
- Remote endpoint answers the MCP handshakeNot checked
README
View on GitHubverifyfirst-mcp
An MCP server for what your verification method cannot see.
A failure that crashes is cheap. A failure that reports success is expensive: the screenshot looks fine, the command exits zero, the endpoint returns 200, and the bug ships.
This server exposes verifyfirst.dev — a registry of 48 such failures, organised by the instrument that missed them rather than by the technology involved, because the instrument is what you know before you know the bug.
Install
uvx verifyfirst-mcp # run it directly
pip install verifyfirst-mcp # or install it
Register with Claude Code:
claude mcp add verifyfirst -- uvx verifyfirst-mcp
Or any MCP client:
{
"mcpServers": {
"verifyfirst": { "command": "uvx", "args": ["verifyfirst-mcp"] }
}
}
Tools
| Tool | What it does |
|---|---|
from_symptom(description) |
Start here when something is wrong but you do not know why. Describe what you see — "deploy ran but nothing changed", "200 but the data is wrong" — and get the failures that produce that appearance, each with its check. |
blind_spots(instrument) |
The one to reach for mid-task. What an instrument cannot see, terse. Takes loose names — curl, 200, pgrep, systemctl, stdout, playwright all resolve. |
list_instruments() |
All six instruments. |
get_instrument(id) |
Full detail plus every recorded failure it missed, each with its check. |
search(query) |
Substring search across the registry. |
get_entry(id) |
One entry, e.g. NS-001. |
get_protocol() |
A five-step checklist for claiming work is done. |
Example
> blind_spots("screenshot")
screenshot (A rendered image) is blind to:
- Time. A still frame cannot distinguish 'renders one frame then stops'
from 'renders one frame correctly'.
- Whether scripts ran at all, as opposed to running and producing this.
...
Notes
Zero dependencies — Python standard library only. Works offline from a bundled
copy of the registry; pass --remote to read the live one, which falls back to
the bundle on any failure.
Every entry in the registry is a failure that genuinely occurs, marked
observed (diagnosed first-hand) or documented (primary source cited and the
check reproduced). None are hypothetical.
CC0-1.0 — public domain. Maintained by Zion Labs.
verifyfirst MCP questions
How do I add verifyfirst MCP to Claude Code?
Run claude mcp add verifyfirst -- uvx verifyfirst-mcp in your terminal. Add --scope user to make it available in every project.
How do I add verifyfirst MCP to Cursor?
Open ~/.cursor/mcp.json and add the server under mcpServers, using the Cursor config shown in the install section above.
How can I find failures based on what I am seeing?
Use the from_symptom tool to describe the issue, such as '200 but the data is wrong', to retrieve failures that produce that specific appearance.
Can I use this server offline?
Yes, the server works offline using a bundled copy of the registry and only requires a connection if you pass the --remote flag.
Are the failures in the registry hypothetical?
No, every entry is a genuine failure that has been either observed first-hand or documented with a cited primary source.