mnemiqCAUTION
Open-source text-to-SQL engine you tune and measure on your own database
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
Text-to-SQL you can tune to your database. /NEM-ik/ — the "m" is silent, as in mnemonic.
By [Agentic Fabriq](https://www.agenticfabriq.com) (YC W26) — made with love, from MIT.
An open-source engine that answers natural-language questions over your database, built so that every stage between the question and the SQL is a setting you can read, change, and measure.
Two answers and a refusal. The third question asks for revenue the database does not hold, and the engine says so — naming the columns it would have needed — instead of returning a number that looks right. That distinction is the whole design.
Read more · Product page · Launch article · Technical report (PDF, DOI 10.5281/zenodo.22806094) · Paper (PDF) · beacon — the grader and results tracker
Why mnemiq exists
Every text-to-SQL product has an accuracy number. Almost none of them were measured on a database that looks like yours.
A system can do well on a benchmark and then struggle on your warehouse because the schema is larger, the naming is different, the business definitions live in people's heads, or the configuration that suited the benchmark simply doesn't suit your data. When it underperforms, a closed system gives you no way to find out why, and nothing to change.
So the question worth asking isn't how accurate is it. It's:
How well will this work on my database — and what can I change if it doesn't?
mnemi
f8d278c49a21OBSERVED · 2026-10-08Connect
Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control. Replace the environment placeholders with a token scoped to the least it needs.
claude mcp add mnemiq-workbench --env MNEMIQ_ENRICH_API_KEY=${MNEMIQ_ENRICH_API_KEY} --env MNEMIQ_LLM_API_KEY=${MNEMIQ_LLM_API_KEY} --env MNEMIQ_ORACLE_TEST_ADMIN_PASSWORD=${MNEMIQ_ORACLE_TEST_ADMIN_PASSWORD} --env MNEMIQ_ORACLE_TEST_PASSWORD=${MNEMIQ_ORACLE_TEST_PASSWORD} -- npx -y [email protected]{
"mcpServers": {
"mnemiq-workbench": {
"command": "npx",
"args": [
"-y",
"[email protected]"
],
"env": {
"MNEMIQ_ENRICH_API_KEY": "${MNEMIQ_ENRICH_API_KEY}",
"MNEMIQ_LLM_API_KEY": "${MNEMIQ_LLM_API_KEY}",
"MNEMIQ_ORACLE_TEST_ADMIN_PASSWORD": "${MNEMIQ_ORACLE_TEST_ADMIN_PASSWORD}",
"MNEMIQ_ORACLE_TEST_PASSWORD": "${MNEMIQ_ORACLE_TEST_PASSWORD}"
}
}
}
}Exposed tools (3)
2 read · 0 write · 1 destructive. Blast radius: 1 tool can delete or overwrite — an agent that can be talked into calling a tool can be talked into calling this one.
| Tool | Risk | Description |
|---|---|---|
db_read | read | Answer a natural-language question over the database (read-only). Returns the |
db_write | destructive | Execute a single INSERT/UPDATE/DELETE (read/write governed separately from db_read). |
get_schema | read | List the tables you are allowed to query. |
Trust audit
CAUTIONgrade C · trust 70/100 Install with care. The audit found things worth knowing before you trust its output.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | WARN |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | WARN |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- declared (4 observation(s))
- Network
- declared (7 observation(s))
- Shell
- none-observed
- Dependencies
- not all pinned
- Secrets in source
- found
Findings (21)
new Blob([`${toCsv(preview)}`], { type: "text/csv;charset=utf-8;" });MNEMIQ_PG_DSN: postgresql://postgres:postgres@localhost:5432/acme
return f"postgresql://mnemiq:mnemiq@localhost:{port}/acme"raise RuntimeError("dsn=postgresql://u:p@h/db refused")leaky = 'permission denied for table hr_prod.payroll_salary; postgresql://svc:[email protected]/x'
leaky = 'permission denied for table hr_prod.payroll_salary; postgresql://svc:[email protected]/x'
db_write
key = hashlib.sha1(f"{self._model}\x00{question}\x00{sql}".encode()).hexdigest()oracle_config_dir="/etc/oracle/wallet", oracle_wallet_password="wp")
assert kwargs["config_dir"] == "/etc/oracle/wallet"
"SELECT pg_read_file('/etc/passwd')", # not even a table read"UPDATE claim SET amount = pg_read_file('/etc/passwd') WHERE id = 1",con.execute("CREATE MACRO all_ssns() AS 1") # the exfiltration fixture's shape_s(llm_base_url="http://169.254.169.254/v1", llm_api_key="k").assert_local_only()
assert "169.254.169.254" in str(exc.value)
uv run mnemiq serve --http # http://127.0.0.1:8080
settings = Settings(llm_base_url="http://127.0.0.1:8000/v1", llm_api_key="k", local_only=True)
settings = Settings(llm_base_url="http://127.0.0.1:8000/v1", llm_api_key="k")
settings = Settings(llm_base_url="http://127.0.0.1:8000/v1", llm_api_key="k", local_only=True)
settings = Settings(llm_base_url="http://127.0.0.1:8000/v1", llm_api_key="k")
react, react-dom, react-markdown, remark-gfm, @tailwindcss/vite, @testing-library/dom, @testing-library/jest-dom, @testing-library/react
Gates applied: no_behavioural_pass.
f8d278c49a21full audit observations/trust-audit/mcp-server/agenticfabriq__mnemiq.json · Report an issue / request a re-scanAudit history
Every audit this server has had. A grade with a past is a grade somebody is still checking.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-08 | f8d278c49a21 | CAUTION | C | 70 | first audit |
Questions
What is the mnemiq MCP server?
Open-source text-to-SQL engine you tune and measure on your own database
What tools does mnemiq expose?
3 in total: 2 read-only, 0 that write, and 1 that can delete or overwrite (db_write). Every one is listed on this page with its risk.
Is mnemiq safe to connect to an agent?
With care. The audit graded it C (70/100) and found 21 things worth knowing before you trust this server, listed below with the exact line each was found on. Separately from the audit: 1 of its tools can destroy data, so scope the token you give it to what you actually need.
What credentials does mnemiq need?
It reads MNEMIQ_ENRICH_API_KEY, MNEMIQ_LLM_API_KEY, MNEMIQ_ORACLE_TEST_ADMIN_PASSWORD, MNEMIQ_ORACLE_TEST_PASSWORD, MNEMIQ_VERIFY_API_KEY and ORACLE_SYS_PASSWORD from the environment. Give it a token scoped to the least it needs — an agent that can be talked into calling a tool can be talked into calling it with your credentials.
How does mnemiq run?
It speaks stdio, so it runs as a local process your client starts. It is published on npm as mnemiq-workbench at 0.1.0.
How current is this page?
The grade is for one exact copy of the source (f8d278c49a21), read on 2026-10-08. The repository is watched and re-audited when it changes.