Skip to content
  1. Skills at a glance
  2. Design and review
  3. /pstack:blast-radius

/pstack:blast-radius

Call when needed

In one lineWhat else could a small change break? It finds out, and runs code to prove why it's safe.

When to use
The change is small but you're not fully at ease with it.
Install with
arena, how, why, unslop
Say it like this
/pstack:blast-radius What else does this change affect?

Just asking vs. using it

The same request, typed plainly or with /pstack:blast-radius. Here's what changes.

Just askingWith /pstack:blast-radius
Hands back a writeup that sounds convincing.Finds the one or two facts safety depends on and proves them by running code.
Greps the callers and calls it done.Reads the library source and version, and follows data and timing grep can't see.
Lists a long string of maybes.Gives each risk a file:line, likelihood, cost, and a way to check it.

What the document says

3 sections, about 650 words, 3 min to read. Click a section to read it.

OpeningIt says the skill finds what a change breaks elsewhere, pairs with how and why, and looks past what grep shows.

  1. 1Don't trust your own writeupGrades each key fact on five levels, from "you said so" to "reproduced in the running app", and says where it stopped.
  2. 2StepsReads the change, finds the one safety fact, looks past grep, weighs risks, proves it by script. Big changes go to arena.
  3. 3What to hand backReturns five parts: what it does, the safety fact and its proof, risks, what was cleared, and the cheapest pre-merge test.

See the wording worth borrowingRead the full text

Wording worth borrowing

A few sentences from the original: what each one makes the AI do, why it works, and how you can borrow it.

  1. Say sounding right counts for nothing

    A blast-radius writeup that sounds right is worthless.

    Makes the AI
    It doesn't hand back a convincing essay; it runs code to prove the point that matters.
    Why it works
    Saying "sounds right" and "is right" are different stops the AI from treating smooth prose as the result.
    Borrow it
    When you ask for a check, add: "However convincing it reads, it doesn't count. Show me a result I can open."
  2. Grade certainty in levels

    For each fact the change's safety depends on, get it as far down this list as is cheap, and say where it stopped.

    Makes the AI
    For each key fact, it tells you how sure it got: just said so, pointed at the code, or actually ran it.
    Why it works
    A ladder from weak to strong lets the AI report where it stopped instead of a vague "looks fine".
    Borrow it
    Ask for a level on each claim: "Mark each one 1 heard it, 2 read the source, 3 tried it myself."
  3. Find the one fact first

    Find the one fact it's safe because of.

    Makes the AI
    Instead of a long list of maybes, it first finds the single fact that makes the change safe.
    Why it works
    Focusing on one fact costs less than listing every possibility, and the check goes deeper.
    Borrow it
    Ask: "Which one assumption does this plan rely on most? Check that first."
  4. Rule out the easy way out

    Listing the callers is not the job.

    Makes the AI
    It doesn't hand back a list of callers; it looks for problems a search can't show.
    Why it works
    Naming the lazy answer the AI would reach for, then the real ask, leaves it no shortcut.
    Borrow it
    Rule out the shortcut first: "Don't just list search results. I want the problems you only find by reading."

Full text

The SKILL.md the AI reads, as written.

descriptionFind what a change could break somewhere else before it ships, beyond the diff, and prove the one fact it's safe because of by running real code instead of writing it up. Use for 'blast radius of X', 'what could this break', or reviewing a small diff you don't trust.

Blast radius

Find what a change breaks somewhere else, before it ships. Use for "blast radius of X", "what could this break", or reviewing a small diff you don't trust yet.

Companion to how and why. how tells you what the code does. why tells you why it's shaped that way. Blast radius tells you what it breaks somewhere else.

Listing the callers is not the job. The agent can grep those in a second. The job is the breakage grep won't show you.

Don't trust your own writeup

A blast-radius writeup that sounds right is worthless. It reads as convincing whether or not it's true. So don't hand back the writeup. Find the one or two facts the whole thing depends on and prove them by running code.

How sure are you

For each fact the change's safety depends on, get it as far down this list as is cheap, and say where it stopped.

  1. You said so. Worthless on its own.
  2. You pointed at the line. A real file:line, or the library's own source.
  3. You showed the bad case can't happen. You walked the failure step by step and it doesn't reach.
  4. You ran it. A script or test that calls the real code and fails loud if you're wrong.
  5. You reproduced it in the running app.

Step 4 is usually one small script that imports the same library the app ships and calls the exact function you're worried about.

Steps

  1. Read the change. The diff, the symbols it adds, changes, and deletes, and what it now does differently, including the part the diff doesn't spell out. Use why step 2 to pull the PR and commits.
  2. Find the one fact it's safe because of. Most changes that look risky are safe because of a single fact, like "this call only drops already-dead cache entries and does nothing else". Find that fact. If it holds, most risky cases are cleared at once. Spend your time here, not on a long list of maybes.
  3. Look where grep stops. Read the source of the library you call, and check its pinned version and any local patch. Work out when things run: microtasks, unmount and teardown, Solid versus React. Follow what a symbol search misses: the JSON an API returns, a DB column, a wire format, another language reading the same bytes, a feature flag, code three hops downstream.
  4. Be honest about each risk. Give it a real chance of happening and a real cost if it does. Keep the risks you confirmed. List the ones you checked and cleared separately. Same rules as why. Cite a real file:line, a search that finds nothing is still an answer, and never make up a caller or an API.
  5. Prove the one fact. Write a script or test that runs the real code, run it, and paste what happened.
  6. For a big or wide change, run it as an arena. Ask more than one model the same question and merge the answers. Different models catch different real bugs.

What to hand back

  • What it does. What changed, including the part that isn't obvious.
  • The one fact it's safe because of. State it, say which step you got it to, and show the proof. If you couldn't prove it, write unproven.
  • Risks. Each names how it breaks, the file:line, how likely and how bad, and how to check. Paste the proof for the ones that matter.
  • Cleared. What you checked and why it's fine.
  • Before you merge. The cheapest test or repro that catches the real bug, including the script you wrote.

Write it through unslop, cite real code, and strip anything private before it goes anywhere public.

Reply: the writeup above, with the one safety fact either proven or marked unproven.

Original by Lauren Tan (poteto), MIT license. Checked against the pstack mirror at 97de40d. View on GitHub

An unofficial beginner's guide to pstack, not its official docs. pstack is made by poteto; the official repo is the source of truth for commands and behavior. The original guide source is on GitHub.

Hosted by airali. Adapted from Xiaoxiaodong's MIT-licensed guide.