Free sample: content calendar cleanup kit
The files that go with the guide Clean a content calendar CSV in n8n without losing campaign IDs: the n8n workflow, a checker, synthetic test files and the expected results.
Download
- File
p1-csv-calendar-cleanup-v0.1.0.zip- Version
- Kit 0.1.0, workflow contract 0.2.0
- Size
- 58,252 bytes. About 4 MB once unzipped: one folder with 44 files.
- License
- MIT, for the Tracetern-original material in the kit
- Data
- Synthetic test files only. No real exports, no customer or personal data.
- SHA-256
9decc839199a45b6857396e510e45e6812c3ab09a9a0dfe13137bb4d01b13f4f
No signup is needed. To check the file on Linux, run sha256sum p1-csv-calendar-cleanup-v0.1.0.zip and compare the result with the SHA-256 above.
What is in the kit
- The n8n workflow from the guide, as a JSON file to import into n8n. It follows workflow contract 0.2.0.
- The contract (
SPEC.md): the input it accepts, the cleanup it may do, the output format and the error codes. - 15 synthetic CSV test files, with a README that explains what each one tests. Ten of them are invalid on purpose.
- 20 expected results, written by hand from the contract: 15 JSON files and 5 cleaned CSV files. The largest CSV was generated by a small script from hand-written definitions. They are not n8n outputs.
- A checker script that judges a result against the contract. It needs Python 3.13, uses only the Python standard library, and reads files without changing them.
- A manifest that lists the 15 test cases and the known limitations. The checker reads it.
- A README with the setup steps, a summary of what was tested and what was not (
RESULTS.md), the license file, and a checksum file for every other file.
What is not in the kit
- Our private test records, such as logs, databases, screenshots and copied outputs.
- The tooling we use to run n8n in isolation, and the tool we use to build and check the kit.
- The checker's own unit tests, and the good and bad sample results used to test it.
- n8n, Node.js and Python. You install them yourself; each has its own license.
How it was tested
On 2026-10-13 (UTC) an AI agent ran the workflow in n8n 2.42.7: three times from the command line on all 15 test files, and in the editor on three of them, with a script operating the editor in headless Chromium 141. The results passed the checker, and the invalid files were rejected as designed. The workflow file in the kit differs from the tested file only in its note text and two comment lines, and it was not run again.
On 2026-10-14 and 2026-10-18 (UTC) the kit itself was checked, also by an AI agent: the ZIP was built twice with identical bytes, unzipped and compared with its checksum file, and the checker agreed with all 15 expected results. These are package checks, not new n8n runs. The kit'sRESULTS.md has the details.
What you need
- A self-hosted n8n that you can open in a browser. We tested only n8n 2.42.7, installed with npm, on Node.js 24.21.0, on Linux.
- Access to the files of the machine that runs n8n, as the guide explains.
- Python 3.13 for the checker. The checker refuses other Python versions.
Not tested
- Docker, n8n Cloud, Windows, macOS, other n8n or Node.js versions, desktop browsers, a person using the editor, an instance that already has workflows, and real calendar exports.
- Unzipping the kit on Windows or macOS. We unzipped it only on Linux.
- Opening the cleaned CSV in a spreadsheet app.
License
The Tracetern-original material in the kit (the workflow, the checker, the synthetic test files, the expected results and the kit documents) is under the MIT License, in the kit'sLICENSE file. It does not cover n8n, Node.js, Python or any other third-party software, and none of them is included.