Find CSV rows with no matching key
Paste two CSV tables and return the rows from A whose key is absent from B — or B-only / both directions. Match by header name or column index, use composite keys, and keep all unmatched duplicates.
About this tool
CSV anti-join is the spreadsheet version of "show me what is missing." Paste two CSV tables, choose the key column, and the tool returns rows from CSV A whose key does not appear in CSV B. That is useful for reconciling exports, finding customers not present in another system, locating inventory codes missing from a reference list, or preparing a cleanup list before a migration.
Use A only for the standard left anti-join. Switch to B only for the reverse check, or
Both sides to emit a symmetric difference with a leading _source column that marks whether
each unmatched row came from A or B. Keys can be header names (id) or 1-based column indexes
(1), and comma-separated keys (first,last) form a composite match. If the key columns have
different names in the two CSVs, set the B key separately — matching is by value, not by header.
Duplicate keys in the lookup table do not multiply rows. If A contains two unmatched rows with the same key, both A rows are kept, because they are real rows missing from B. You can choose comma, tab, semicolon, pipe, or any single-character delimiter, and optionally make key matching case-insensitive or trim whitespace around key values.
Worked example
CSV A:
id,name
1,Alice
2,Bob
3,Carol
CSV B:
id,city
2,Berlin
3,Cairo
4,Delhi
Key: id, direction: A only.
Output:
id,name
1,Alice
FAQ
What is an anti-join?
An anti-join returns rows from one table when their key does not exist in another table. It is the
opposite of an inner join: instead of keeping matches, it keeps non-matches. In SQL terms, this is
similar to LEFT JOIN ... WHERE right.key IS NULL or a NOT EXISTS query.
Can the key column have a different name in CSV B?
Yes. Put the A key in Key column(s) in A and the B key in Key column(s) in B. For example,
A might use customer_id while B uses id. The tool compares the key values, so the header names
do not need to match. Leave the B key blank when both files use the same key name or index.
How are duplicate keys handled?
The comparison side is treated as a key set, so duplicate keys in B do not multiply or otherwise
affect A's output. On the output side, every unmatched row is kept. If CSV A has two rows with key
x and x is absent from B, both rows appear in the result.
Can I compare on more than one column?
Yes. Use a comma-separated list such as first,last or 1,3. The same number of key columns must
be supplied for CSV B. A row matches only when all key parts match after the case-sensitivity and
trim settings are applied.
Does the tool upload my CSV data?
No. The anti-join runs in WebAssembly inside your browser tab. Your pasted CSV text is processed locally and is not sent to a server.
Developer & Automation Access
Run it from the terminal
Same engine as this page, headless — via the gizza CLI:
gizza tool csv-anti-join "id,name
1,Alice
2,Bob
3,Carol" 'b=id,city
2,Berlin
3,Cairo
4,Delhi' 'key=id'New to the CLI? Get gizza →
Open it by URL
Pre-fill and auto-run this tool with query parameters — the names match the API/CLI:
https://gizza.ai/tools/csv-anti-join/?a=id%2Cname%0A1%2CAlice%0A2%2CBob%0A3%2CCarol&b=id%2Ccity%0A2%2CBerlin%0A3%2CCairo%0A4%2CDelhi&key=id&key_b=id&direction=a-only&delimiter=%2C&case_sensitive=true&trim_keys=trueMachine-readable descriptor: tool.json — title + parameters JSON Schema for agents.