Agent skill

Skill Pressure Test

by nyldn in nyldn/claude-octopus

Interrogate a plan, decision, or design one question at a time until it holds — use to stress-test your own thinking before committing to it

MITAuto-check passedAgent Workflows

Install Skill Pressure Test

skills CLI
$ npx skills add nyldn/claude-octopus --skill skill-pressure-test -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install nyldn/claude-octopus skill-pressure-test --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/nyldn/claude-octopus.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/skill-pressure-test .claude/skills/skill-pressure-test && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
skill-pressure-test
GitHub stars
4.2k
Used in
1 other repo
Token cost
~1.4k tokens
SKILL.md length
869 words
Files
2
Skills in repo
62
Repo updated
First seen
Licence
MIT

At a glance

Interrogate a plan, decision, or design one question at a time until it holds — use to stress-test your own thinking before committing to it

  • Works in 3 steps: The repository — code, config, history,… → The user — for every decision,… → External sources only when the decision…
  • Stress-test your own thinking before committing to it
  • SKILL.md covers When To Use, When Not To Use, Inputs and Workflow, plus 4 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Skill Pressure Test is an agent skill from nyldn/claude-octopus. Interrogate a plan, decision, or design one question at a time until it holds — use to stress-test your own thinking before committing to it

Its SKILL.md is about 1.4k tokens, which your agent loads only when the skill is triggered. The skill folder holds 2 other files (for example `agents/openai.yaml`).

It sits in Agent Workflows, covering Load testing. The repository describes itself as: Run multiple AI models against the same research, design, or coding task. Surface disagreements before you ship. The licence is MIT.

When your agent uses it

  • Stress-test your own thinking before committing to it
  • Tasks that involve Load testing

Example prompts

  • “/skill-pressure-test”

Workflow steps

3 steps, taken from the first numbered list in SKILL.md.

  1. The repository — code, config, history, tests. Facts live here.
  2. The user — for every decision, preference, and priority.
  3. External sources only when the decision turns on something outside the repo,

What it can do on your machine

Read from SKILL.md and the folder at commit e14b84f. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    Links to these hosts (documentation or services it may open):

    • github.com

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Skill Pressure Test loads about 1.4k tokens when it runs. Until then it costs about 40 tokens; SKILL.md has 869 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~40
When it runs · the whole SKILL.md, loaded when a task matches
~1.4k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from nyldn/claude-octopus at commit e14b84f, republished under its MIT licence (© nyldn). 869 words, ~1,407 tokens.

Download SKILL.mdSave it as .claude/skills/skill-pressure-test/SKILL.md (or your agent's skills folder). This skill also uses 1 other file; get the full folder from GitHub.
name
skill-pressure-test
description
Interrogate a plan, decision, or design one question at a time until it holds — use to stress-test your own thinking before committing to it
disable-model-invocation
true

Host: Codex CLI — This skill was designed for Claude Code and adapted for Codex. Cross-reference commands use installed skill names in Codex rather than /octo:* slash commands. Use the active Codex shell and subagent tools. Do not claim a provider, model, or host subagent is available until the current session exposes it. For host tool equivalents, see skills/blocks/codex-host-adapter.md.

Pressure Test

Interview the user relentlessly about a plan, decision, or design until you both reach a shared understanding of it. Walk each branch of the decision tree, resolving dependencies between decisions one at a time.

This is not brainstorming. skill-thought-partner opens a space up and looks for what might be there; this closes one down and looks for what is wrong with it. Use this when there is already a position on the table and the risk is that it is wrong in a way nobody has said out loud.

Adapted from the grilling skill in mattpocock/skills (MIT).

When To Use

  • The user has a plan, spec, or design and wants it attacked before they commit.
  • A decision keeps getting deferred because its dependencies are tangled.
  • Scope feels slippery and nobody can say precisely what is in it.
  • Before a large or hard-to-reverse change, where being wrong is expensive.

When Not To Use

  • To generate options. That is skill-decision-support.
  • To explore an open space with no position yet. That is skill-thought-partner.
  • When the user wants the work done, not examined. Say so and stop.
  • When you can settle the question by reading the codebase. Read it instead.

Inputs

The plan, decision, or design under test, in whatever form exists — a document, a paragraph, or just the last few turns of conversation. Nothing needs writing up first; extracting the shape is part of the job.

Workflow

Ask one question at a time. Wait for the answer before asking the next. A batch of questions is a questionnaire, and it gets questionnaire answers: shallow, and shaped by whichever one the reader happened to care about. One question, answered properly, changes what the next question should be.

Carry a recommended answer with every question. "What should happen when the token expires?" is work handed back. "What should happen when the token expires? I would refresh silently and only surface an error if the refresh fails, because the alternative interrupts the user mid-task — do you agree?" is a question that can be answered in one word, and disagreed with precisely.

Look facts up; put decisions to the user. If the answer is discoverable in the filesystem, the git history, a config file, or a tool you can run, find it — asking is a tax on the user for work you could have done. Decisions are different: they are the user's, and no amount of reading the codebase produces them. Do not infer a decision from a pattern and proceed as though it were settled.

Follow dependencies, not a list. When an answer makes another question moot, drop it. When it opens two new ones, ask those before returning. The order is whatever the tree dictates.

Represent each material decision with its evidence, dependencies, owner, resolution, and the work it unblocks. Facts may be researched autonomously. Preferences and new permissions remain human decisions. Do not answer those on the user's behalf to keep an unattended run moving. If dependencies form a cycle, name it and withhold ready status until the decisions are recut.

Name disagreement when you find it. If an answer contradicts an earlier one, say which two and ask which holds. Quiet reconciliation is how a plan ends up meaning two things.

Show full SKILL.md (271 more words)Show less

Provider Or Data Priority

  1. The repository — code, config, history, tests. Facts live here.
  2. The user — for every decision, preference, and priority.
  3. External sources only when the decision turns on something outside the repo, and say so when it does.

Stop Or Checkpoint Rules

  • Do not act on the plan until the user confirms you have reached shared understanding. This skill produces agreement, not changes.
  • Stop when the remaining questions no longer change what anyone would do. More interrogation past that point is theatre.
  • Stop and say so if the plan turns out to be sound. Manufacturing an objection to look rigorous is worse than finding none.
  • If an answer invalidates the premise of the whole plan, stop the interview and raise that. Do not keep grilling the details of something already dead.
  • If new evidence invalidates a resolved decision, reopen its dependent work and record the reason. A completed task does not make stale guidance true.

Output Contract

  1. Where it stands — sound as-is, sound with the changes below, or unsound.
  2. Decisions settled — one line each, in the order they were made, since later ones often depend on earlier ones.
  3. Changes to the plan — what the answers actually altered.
  4. Still open — questions raised and not resolved, and what each one blocks.
  5. Facts found — anything looked up that the user did not know, with where it came from.

Verification

  • Every question was asked singly and answered before the next.
  • No decision in the summary was inferred rather than answered.
  • Every fact cites where it came from.
  • Nothing in the plan was changed during the interview.

© nyldn, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 1 other file in skills/skill-pressure-test of nyldn/claude-octopus.

  • SKILL.md
  • agents/openai.yaml

Open the folder on GitHubat commit e14b84f

Used in 1 other repository

We found 2 copies of this SKILL.md (exact, near-identical or edited) in other folders, from 1 other GitHub owner. This page covers the copy in nyldn/claude-octopus, which our catalogue first saw on October 7, 2026.

Compare with similar skills

Skill Pressure Test next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Skill Pressure Test compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Skill Pressure Test this skillnyldn/claude-octopus4.2k1 repos~1.4kAutomated safety check: PassMIT
Grillingpietheinstrengholt/rssmonster56431 repos~510Automated safety check: PassMIT
Idea Genieboshu2/agentops4481 repos~1.8kAutomated safety check: PassApache-2.0
Adversarial ReviewDanMcInerney/architect-loop626—~1.1kAutomated safety check: PassMIT
Deep Divebyungjunjang/jangpm-meta-skills120—~2.5kAutomated safety check: PassNone
Grillingopencrvs/opencrvs-core1202 repos~205Automated safety check: PassCustom licence

Similar skills

  • Grilling

    pietheinstrengholt/rssmonster

    Grill the user relentlessly about a plan, decision, or idea.

    564 GitHub starsUsed in 31 repos~510 tokens
    Agent WorkflowsAuto-check passed
  • Idea Genie

    boshu2/agentops

    Brainstorm evidence-backed options for what to build, or stress-test an idea.

    448 GitHub starsUsed in 1 repo~1.8k tokens
    Agent WorkflowsAuto-check passed
  • Adversarial Review

    DanMcInerney/architect-loop

    A skill your agent uses when the architect factory orchestrator dispatches a fresh strategist subagent to harden a draft spec: falsify it with file:line evidence, fold the surviving findings into a…

    626 GitHub stars~1.1k tokensUpdated 26 days ago
    Agent WorkflowsAuto-check passed
  • Deep Dive

    byungjunjang/jangpm-meta-skills

    Socratic interview skill to deepen a spec or refine an existing agent blueprint.

    120 GitHub stars~2.5k tokensUpdated 2 mo ago
    Agent WorkflowsAuto-check passed
  • Grilling

    opencrvs/opencrvs-core

    Grill the user relentlessly about a plan or design. An agent skill from opencrvs/opencrvs-core.

    120 GitHub starsUsed in 2 repos~205 tokens
    Agent WorkflowsAuto-check passed
  • Deep Dive

    byungjunjang/jangpm-meta-skills

    A Codex skill for Socratic interviews that deepen a spec or refine an existing agent blueprint.

    120 GitHub stars~2.7k tokensUpdated 2 mo ago
    Agent WorkflowsAuto-check passed

More from nyldn/claude-octopus

All 62 skills in this repo
  • Octopus Quick

    nyldn/claude-octopus

    Quick execution for ad-hoc tasks without full workflow overhead — use for small, self-contained requests

    4.2k GitHub starsUsed in 1 repo~2.2k tokens
    Auto-check passed
  • Octopus Research

    nyldn/claude-octopus

    Thorough research across multiple sources — use for complex topics needing broad synthesis

    4.2k GitHub starsUsed in 1 repo~1.9k tokens
    Auto-check passed
  • Octopus Security Audit

    nyldn/claude-octopus

    OWASP compliance, vulnerability scanning, and adversarial red team testing — use for security reviews

    4.2k GitHub starsUsed in 1 repo~2.3k tokens
    Auto-check passed
  • Skill Audit

    nyldn/claude-octopus

    Audit codebases for quality, consistency, and broken patterns — use for pre-release or tech debt review

    4.2k GitHub starsUsed in 1 repo~3.2k tokens
    Auto-check passed
  • Skill Content Pipeline

    nyldn/claude-octopus

    Extract patterns and anatomy from URLs — use to reverse-engineer content strategies from live pages

    4.2k GitHub starsUsed in 1 repo~3.9k tokens
    Auto-check passed
  • Skill Context Detection

    nyldn/claude-octopus

    Auto-detect work context (Dev vs Knowledge) — use to tailor workflows based on current task type

    4.2k GitHub starsUsed in 1 repo~2.6k tokens
    Auto-check passed

Questions about Skill Pressure Test

What does Skill Pressure Test do?

Interrogate a plan, decision, or design one question at a time until it holds — use to stress-test your own thinking before committing to it. Skill Pressure Test is an agent skill from nyldn/claude-octopus.

When should I use Skill Pressure Test?

Skill Pressure Test fits situations like: stress-test your own thinking before committing to it; tasks that involve Load testing.

How do I install Skill Pressure Test in Claude Code?

Run `npx skills add nyldn/claude-octopus --skill skill-pressure-test -a claude-code`. Or copy the skill folder (skills/skill-pressure-test in nyldn/claude-octopus) into .claude/skills/skill-pressure-test in your project. Claude Code loads it when a task matches its description.

How do I install Skill Pressure Test in Codex?

Run `npx skills add nyldn/claude-octopus --skill skill-pressure-test -a codex`. Or copy the skill folder (skills/skill-pressure-test in nyldn/claude-octopus) into .agents/skills/skill-pressure-test in your project. Codex loads it when a task matches its description.

Can I use Skill Pressure Test in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add nyldn/claude-octopus --skill skill-pressure-test -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/skill-pressure-test, .gemini/skills/skill-pressure-test, .github/skills/skill-pressure-test and .opencode/skills/skill-pressure-test in your project.

What does Skill Pressure Test need to run?

SKILL.md names no scripts, command-line tools or credentials: Skill Pressure Test is instructions for the agent only.

Does Skill Pressure Test access the network?

SKILL.md names 1 domain. As links in the text: github.com. This is read from the text; nothing was executed.

Is Skill Pressure Test safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Skill Pressure Test use?

Skill Pressure Test is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Skill Pressure Test use?

About 1.4k tokens (SKILL.md is roughly 5.6k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Skill Pressure Test?

Skills that share tags, products or a category with Skill Pressure Test: Grilling (pietheinstrengholt/rssmonster, 564 stars), Idea Genie (boshu2/agentops, 448 stars), Adversarial Review (DanMcInerney/architect-loop, 626 stars) and Deep Dive (byungjunjang/jangpm-meta-skills, 120 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Skill Pressure Test?

nyldn (a GitHub user) maintains it in nyldn/claude-octopus, which has 4,192 GitHub stars. The repository holds 62 skills in this directory. The repository was last updated on October 9, 2026.

Source: nyldn/claude-octopus on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.