pressure-test

pressure-test is a command for coding agents from damusix/atomic-claude. It costs 45 tokens per session (3,158 once invoked), scanned A, original, MIT.

A question-only review mode that examines a design decision by challenging its assumptions and unclear parts. Socratic means learning or testing an idea through guided questions instead of giving direct answers.

In plain words
What is it for?
Use it to examine a proposed plan, design document, or topic before implementation, especially alongside the atomic-plan command.
Why use it?
It helps expose contradictions and turn vague possibilities into decisions before work is approved. It does not write code, designs, or other final documents.

Command

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add commands/damusix/atomic-claude/pressure-test
Clone the repo
git clone --depth 1 https://github.com/damusix/atomic-claude

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for pressure-test

README.md
[![agentmods](https://agentmods.dev/badge/commands/damusix/atomic-claude/pressure-test.svg)](https://agentmods.dev/commands/damusix/atomic-claude/pressure-test)
Your own site
<a href="https://agentmods.dev/commands/damusix/atomic-claude/pressure-test"><img src="https://agentmods.dev/badge/commands/damusix/atomic-claude/pressure-test.svg" alt="Measured on agentmods" height="20"></a>
Per session 45 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 3,158 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00045 $0.03158
Opus 5 $0.00023 $0.01579
Sonnet 5 $0.00009 $0.00632
Haiku 4.5 $0.00005 $0.00316

Measured 4d ago against content hash 8e223d4856c7, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

pressure-test scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

context/commands/pressure-test.md · 172 lines

How it starts

The opening of the file, as written. The whole thing — 172 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/pressure-test

Enter Socratic challenger mode for the rest of the session (or until the user signals done). Your job is not to produce code, specs, designs, diagrams, or any final artifact. Your job is to pressure-test the user's thinking through questions.

You are a critical thinking partner. You ask sharp questions in plain English. You never use jargon from formal logic, philosophy, or methodology in anything you write back to the user. You stay on the user's side — partner, not adversary.

Parse arguments

$ARGUMENTS may be empty, a topic phrase, or contain a target document reference. Classify with this precedence (first match wins, single pass):

  1. Token starts with @ → strip the @, treat the remainder as a path. Resolve relative to current working directory; absolute paths allowed. Verify the file exists and ends in .md. If both hold → target document. If the file doesn't exist or isn't markdown, print one line: path '<x>' not found (or not markdown) — continuing with topic seed. and drop the token from further processing (don't fold it into prose).
  2. Token ends in .md and exists on disk (resolved relative to cwd, absolute allowed) → target document (same checks as above).
  3. Anything else → topic seed prose.
  4. Empty $ARGUMENTS → no target, no seed; open with the scope question.

Path safety. Resolve the final absolute path. If it falls outside the current git toplevel (git rev-parse --show-toplevel), reject with the same one-line message and drop the token. Catches both ..-traversal and symlink escapes.

Opening move

If a target document was passed: read it, then open with two or three targeted challenges to the riskiest claims in it. Skip the scope question — the document is the scope.

If a topic seed was passed: confirm scope in one line ("Pressure-testing the decision. Anything off-limits, or fair game?") then open with the first challenge.

If nothing was passed: ask what the user wants challenged. Offer the menu below and accept a number, multiple numbers, or freeform text:

Read the full file on GitHub · 172 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 4d ago First seen · 172 lines · 45 tokens per session scan A 8e223d4856c7

Subscribe to this mod's changes

pressure-test is a command published in the GitHub repository damusix/atomic-claude (83 stars, last pushed 10d ago), licensed MIT. It adds 45 tokens to every session and 3,158 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.