A per-change watch plan template with five rules and a five-check script (free sample)
by PaperChaseWebb
From Release Watch: five rules for watching a change after it goes live, a plan template, one prompt that has your agent write the plan from the diff, and a read-only script that checks the plan before anything watches it, with a worked example.
- Price
- Free
- Difficulty
- Beginner
- Setup
- 10 min
- Format
- .md file
release-watch-plan-template-and-five-checks-free-sample.md · 4 KB · direct download, no signup
What you end up with
Your next change has a written plan for what should not move, with a timezone, a human owner, a persist rule, floors on ratios and no automatic rollback.
What you get
- Five rules for per-change watches
- A plan template
- One prompt for all four harnesses
- watch_mini.py: five read-only checks
- Worked example: 5 findings, then 0
Who it is for
Anyone who merges changes with an agent's help and wants a small habit for watching each one, before buying a runner.
Works with
Solid chips: the agent is named in this listing's own tools list. Dashed chip: Plain Markdown, so it also pastes into other agents such as Grok Bot, Hermes, OpenClaw or Muse. We list those as compatible by format only, not as tested.
Tools needed
Cost to run: Free.
Time to result: About ten minutes to a checked plan
Sample prompt from the pack
Read the diff for this change and write watch.plan.json from the template in this file. Run python3 watch_mini.py watch.plan.json until it prints 0 findings, then show me the plan. Do not start any watch.
The rules and the template are in the file.
Free preview of the file
# Release Watch, free sample: a watch plan template, five rules and a five-check script
Version 1.0, 2026-10-28. By PaperChaseWebb. Works with Grok Bot, Claude Code, Codex and Hermes. Python 3 standard library. Tested on Linux with Python 3.13.5 as described below.
After a change goes live, most teams either watch nothing in particular or ask an agent to stare at dashboards. This sample gives you the smaller, cheaper habit: write down, per change, what should not move, then check that plan before anything watches it. Credit for the core idea, an agent writing the spec and a fixed script running it, goes to Ben Congdon's post on agentic monitoring: https://benjamincongdon.me/blog/2026/10/19/Agentic-Monitoring/
## Five rules
1. One plan per change, written from the diff. Each signal says which changed paths it covers and why.
2. Every time carries a timezone, and the baseline window is at least as long as the watch window.
3. Trip only after two or more bad windows in a row. One blip is not an incident.
4. Ratio rules need a floor, so near-zero traffic does not page you. Watch for drops, not only spikes.
5. Nothing runs automatically. The watch proposes; a named person decides any rollback.
## A plan template
```json
{
"change_id": "rel-coupons",
"change_time": "2026-10-20T12:00:00-10:00",
"owner": "Dana Ops",
"signals": [
{"name": "checkout errors", "rule": "increase", "threshold": 2.0, "persist": 2, "floor": 0.02},
{"name": "checkout traffic", "rule": "drop", "tWorth knowing
- The sample checks a plan; it does not run one. The runner, Prometheus source, triage prompt, approvals, fifteen-check checker and eval set are in the full pack.
- On 2026-10-28 watch_mini.py gave 5 findings on the worked example, then 0 on the template, and exit 2 on a missing file. The prompt was not run inside a harness.
Questions
Why write a plan per change?
A global alert does not know what the change touched. A plan tied to the diff watches the signals that change could move, so it is quieter and easier to act on.
What is in the full pack?
A runner that reads CSV or Prometheus read-only with increase, above and drop rules, plan and triage prompts, skill files for Claude Code, Codex and Hermes and a Grok Bot rules block, approvals tied to the plan hash, fifteen checks with a fixture, and a labeled eval set. $24.
Details
Version 1.0 · Updated Oct 28, 2026 · Markdown version of this page
License: Personal and commercial use. Download terms
Related
Release Watch: a per-change monitoring plan your agent writes from the diff, run by a fixed script
After a change goes live, your agent reads the diff and writes a small watch plan for that change: which signals it could move, against which baseline, by how much, for how long. A fixed script runs the plan every five minutes, so no agent sits in a loop. On a trip, a second agent triages read-only and proposes one next step. Rollbacks, flag flips and alert changes stay with you.
The builder and tester contract and a red-then-green check (free sample)
From Two-Agent Build and Test Loop for paid apps: the six rules that keep a coding agent from grading its own work (two sessions, the tester never edits code, the builder never touches tests or says passed, red first, a script runs the tests, sandbox only with owner yeses on paid apps), a story card, and a read-only Python check on your run's ledger, with a worked example and its expected output.
The planner and worker contract and a plan preflight check (free sample)
From Planner Over Cheap Models: the six rules that let a strong model plan while cheap or local models write the code (planner never edits, a check on every task, the check runs instead of the worker, parallel tasks own different files, high risk goes up a tier, owner yes before any worker starts), plus a read-only Python check on your plan, with a worked example and its expected output.