Scanisaur
The apex predator of agent-generated SQL.
Scanisaur is an MCP server that AI agents call before they run SQL. It reads only warehouse metadata (tables, columns, types, partitioning, clustering, row counts and sizes) and uses it to:
- validate the query: are these tables and columns real?
- optimize it: catch missing partition filters,
SELECT *on wide tables, joins with no condition and other costly patterns, each with a concrete fix; - quantify it: estimate bytes scanned and cost before anything runs.
Scanisaur never reads table data and never runs the queries it checks. The agent's own SQL tool does that.
Status: pre-alpha. Nothing is published yet. BigQuery is the first supported warehouse; Snowflake follows.
Planned MCP tools
| Tool | Purpose |
|---|---|
scanisaur_schema_search |
Find relevant tables and columns by keyword |
scanisaur_schema_describe |
Compact, token-efficient table descriptions with partition and cluster keys |
scanisaur_check_sql |
Verdict (pass, warn, block), cost estimate, findings with fixes, and a tracking tag |
Development
You need uv.
uv sync
uv run scanisaur --version
uv run pytest
See CONTRIBUTING.md for the full set of checks, and docs/adr/ for the design decisions so far.
License
1 maintainer
Metadata
1 maintainer
Release files for scanisaur 0.0.1
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| scanisaur-0.0.1.tar.gz | 77.1 kB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| scanisaur-0.0.1-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 89.9 kB
Release files / scanisaur-0.0.1.tar.gz
| Download URL | scanisaur-0.0.1.tar.gz |
|---|---|
| Size | 77.1 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
61784f1a57653d6d1787d17c0963edb6ee23be47bd0c86658d118564895aee88
|
|
BLAKE2b-256 checksum How to use checksums |
6e765a871bd757e25938f8fa28266c79c867c56960f7f12cd703ec8be8933cee
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Oct 1, 2026.
Transparency logRelease files / scanisaur-0.0.1-py3-none-any.whl
| Download URL | scanisaur-0.0.1-py3-none-any.whl |
|---|---|
| Size | 12.8 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
f973082554fe7ad9d0275aab64d0e17470a8c9f86121ee439f6846b99c6edd6e
|
|
BLAKE2b-256 checksum How to use checksums |
9c6e722699f1e47a288485c3b92777cf476d7691f387f9f1977e684f2ca87d54
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Oct 1, 2026.
Transparency log
