- Home
- Skills
- Code Quality & Review
- Pre-Merge Checklist
Pre-Merge Checklist
Machine checks before you merge: clean tree, diff size, secrets, markers, tests. Each check reported as run or not run.
$60
Works with the AI Agents you already use
Pre-Merge Checklist
Example session with this skill installed
I am about to merge the branch feature-config into main. Can you look it over and tell me if it is ready? Tests run with: python3 -m unittest discover -s tests
- Read your context and instructions
- Compiled the pre-merge checklist
Summary
The agent ran the checks, read the diff, and reported each item as RAN, NOT RUN or READ. Verdict: NOT READY. The output below is from running the script on a small test branch.
Script output
$ python3 scripts/premerge_check.py --test-cmd "python3 -m unittest discover -s tests"
PASS clean-tree
PASS diff-size: 2 files, 5 changed lines
FAIL secrets: assigned secret at config.py:2
PASS conflict-markers
WARN todo-markers: config.py:1
PASS binary-files
WARN tests-touched: 2 code file(s) changed, no test file changed
PASS tests: exit 0
Exit code 1, because of the FAIL line. The secret value is never printed.
Report
| State | Check | Result |
|---|---|---|
| RAN | clean-tree | PASS |
| RAN | diff-size | PASS, 2 files and 5 changed lines |
| RAN | secrets | FAIL, assigned secret at config.py:2 |
| RAN | todo-markers | WARN, config.py:1 |
| RAN | tests | PASS, exit 0 |
| NOT RUN | banned text | no --ban text was given |
| READ | diff | sub() in app.py has no test. The passing suite says nothing about it. |
Verdict
NOT READY. Remove the key from config.py and read it from the environment. If the key is real, rotate it, because it stays in the commit history. Then add a test for sub(). The agent did not merge.
Connects securely to your tools. The creator never sees your data.
About this skill
Agents say a branch is ready. This skill makes them show it.
The agent runs a script on your branch and reports one line per check:
- clean working tree
- diff size against limits you can set
- secret-like values in added lines (the value is never printed, only the kind and the place)
- merge conflict markers
- TODO and FIXME markers (warning)
- text you ban with --ban
- binary files (warning)
- code changed with no test changed (warning)
- your test command, if you give one
Then the agent reads the diff itself and answers four questions: did it do what the task asked and nothing else, does new behaviour have a test, are there files the task never mentioned, does any text claim more than the code shows.
The test command runs through the shell, and the check fails if there is nothing to review. The report format has three states: RAN, NOT RUN, READ. A check that did not run is never reported as passed. The skill ends with READY, NOT READY or NEEDS OWNER, and the agent does not merge.
Standard library only. Written for Python 3.10 or later, tested on 3.13 only. The script has 19 tests.
What it does not do It does not judge whether the code is correct, does not find every secret, does not review design, and cannot run tests it is not told about. A branch can pass every line and still be wrong. It does not merge.
Before you use this These skills are new and have not been tested widely. They come from a project that works and that I use. You are responsible for what you run and for reviewing what your agent does with them, so check results before you rely on them. The author is not liable for loss or damage, in line with the MIT licence.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Code Quality & Review.
Requires Python 3 and Git. Cannot autonomously fix failures. Size limits default to 30 files or 800 lines unless overridden.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed today
- Software
Git
Python 3
Permissions
No network access. The script runs git commands and reads the diff. It writes no file itself. Your own test command runs through your shell and can do whatever your tests do.