- Home
- Skills
- Productivity
- Plain Change Report: what your AI changed, in plain words
More screenshots
Works with the AI Agents you already use
Plain Change Report: what your AI changed, in plain words
Turns finished coding work into a short report for an owner who does not read code: what is different, how to see it, what was and was not checked, and the one decision left.
$5
Plain Change Report: what your AI changed, in plain words
Example session with this skill installed
The AI just finished this change to my notes app. My request was: "When I pick dark, it should still be dark next time I open it." I don't read code. Tell me what changed, in words I can act on.
- Read your context and instructions
- Compiled the plain change report
The notes app now remembers your look: pick Dark once and it stays dark.
What is different for you
- Choose Dark, close the app, and it opens dark next time.
- Someone opening it for the first time still starts on Light.
See it yourself
- Open the notes app.
- Under Look, choose Dark.
- Close it and open it again. It is still dark.
Checked, and not checked
- Checked: 3 automatic tests pass. The app was opened, set to Dark, closed and reopened: still dark. A new visitor got Light. The AI also fixed the command that runs the tests, which had stopped working.
- Not checked: phones and Safari. Neither could be opened here.
Needs you
Should a first-time visitor start in their device's own look (dark or light), or always in Light?
What the report was built from (recorded in the run): two commits (the change, and a fix to the test command), npm test failing before that fix and passing 3 of 3 after it, and the app opened, set to Dark, closed and reopened. A real run on a practice notes app, 23 Oct 2026.
Connects securely to your tools. The creator never sees your data.
About this skill
If you direct software but do not read code, most end-of-task reports are file names and tool talk. Plain Change Report has your agent gather the facts first (your request, the actual changes, what was tested and how), then report in five parts: the outcome in one line, what is different for you, numbered steps to see it yourself, what was checked and what was not, and the one thing that needs you. Every claim has to come from something that was run or seen, so "should work" never passes for "done". It carries a plain-words table for the usual jargon, and it credits work to whoever actually did it.
Created by Greygray, with AI collaboration openly included in the process.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Productivity.
It reports; it does not test. It can only say "checked" about what was actually run or seen during the work, so if nothing was tested, the report says so. It reads the changes from git, so work outside a git project gives it less to go on. A change with nothing to see (speed, safety, tidying) gets one line saying so and what it protects you from, not steps to see it.
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 29 days ago
- Software
Git
Permissions
Terminal / Shell: to run read-only git commands (git status, git diff, git log) and see what changed. Read Files: to read the changes and any test results.