agentleFS
Sign inStart free
davidondrejdavidondrej/cloudroom/skills/risky-changes/SKILL.md

risky-changes

Skill1.5k starsChanged 6 days agoApache-2.0View on GitHub

Verify assumptions before implementing large or risky changes to APIs, provider data, billing, pricing, quotas, or defaults. Use when a mistake could affect customers or the user asks if a change is safe to ship. Checks real-world impact beyond passing tests.

2 other projects use a version of this.

Adopted by

Steady: 0 new projects took it in the last 4 weeks, none the 4 weeks before.thin data

Week of Sep 13: 0 new projectsWeek of Sep 20: 0 new projectsWeek of Sep 27: 0 new projectsWeek of Oct 4: 0 new projectsWeek of Oct 11: 0 new projectsWeek of Oct 18: 0 new projectsWeek of Oct 25: 0 new projectsWeek of Nov 1: 0 new projects
Sep 2026Oct 2026Nov 2026
New projects taking a version of this file each week, over the 8 weeks our latest run read: only 0 in all, so this is a hint, not a trend.

Most recently taken up by

Who's talking about davidondrej/cloudroom

The biggest 5 of 25 mentions linking to the repository that our crawl found. Each line is their own headline.

See all 25

Talked about on GitHub (25 times).

On GitHub · 25

20 more on GitHub
  • Other versions2 projects use a version of this.

    Word for word · 2

    1. davidondrej/skills★ 4.3k100% same
    2. davidondrej/cloudroom-core★ 1.5k100% same

    Edited · 0

    None found.

    Wrote their own version · 0

    None found.

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository8 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

7 rules. About 0 are common, 7 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈7

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use for changes where being wrong is expensive or customer-visible:

    line 12
  2. Distinctive:

    If unsure whether a change qualifies, apply this process.

    line 20
  3. Distinctive:

    Use available research tools and reliable sources.

    line 28
  4. Distinctive:

    If evidence contradicts an assumption, reconsider the design before coding.

    line 34
  5. Distinctive:

    Run 10–20+ realistic cases against the real endpoint or provider:

    line 38
  6. Distinctive:

    Use hard numbers where possible.

    line 42
  7. Distinctive:

    Read-only production analysis also counts as measurement.

    line 44

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Risky Changes

A filter once passed every test but disabled ~99% of the feature on live data. Tests check code correctness; research and live measurement check whether a change is useful.

When this applies

Use for changes where being wrong is expensive or customer-visible:

  • Public API fields, filters, or response shaping
  • Dropping, transforming, or reordering upstream data
  • Billing, pricing, caps, or quotas
  • Defaults, thresholds, or provider request parameters
  • Unverified assumptions about external data or user behavior

If unsure whether a change qualifies, apply this process.

1. Identify assumptions

List the assumptions the change depends on. Mark which have evidence and which are still unverified.

2. Research before implementing

Use available research tools and reliable sources. Investigate each distinct question separately, covering at least:

  • How do leading products handle this design decision?
  • What does real-world data look like: frequencies, shapes, and edge cases?
  • What do users or agents actually need?

If evidence contradicts an assumption, reconsider the design before coding. If research is unavailable or inconclusive, state the gap; do not treat the assumption as verified. Do not ship while material assumptions remain unverified.

3. Measure real behavior

Run 10–20+ realistic cases against the real endpoint or provider:

  • Base cases on real usage; vary topics, parameters, languages, and edge conditions.
  • Define benchmarks per case: speed, quality, accuracy, and how often the new behavior occurs.
  • Use hard numbers where possible. For subjective quality, use blind, criteria-based judging.
  • Compare before and after when both can be measured.
  • Read-only production analysis also counts as measurement.

Save the cases and results in the project's evals folder, e.g. docs/evals/YYYY-MM-DD-<endpoint>-<focus>.md. Create the folder if needed. Without this record, the change is not verified. Unit tests do not replace live measurement.

4. Confirm product-owner approval before shipping

Present decisions that affect what customers see or pay, with supporting research and measurements. Get the product owner's approval for those decisions; do not bury them in a plan or code default. Existing approval counts if it covers the actual behavior being shipped.

5. Verify after deployment

Within one day, measure the change on real traffic through read-only production analysis or a live sweep. Report results that differ from expectations immediately.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free