gauntlet: Skill for Claude Code

.claude/skills/writing-gauntlet-stories/SKILL.md

writing-gauntlet-stories is a skill for Claude Code from prime-radiant-inc/gauntlet. It costs 70 tokens per session (2,649 once invoked), scanned A, original, Apache-2.0.

A writing guide for Gauntlet story cards, which describe an outcome, the person or role involved, and observable conditions for success. It distinguishes these cards from Agile user stories, step-by-step test scripts, and BDD scenarios.

In plain words
What is it for?
Use it to write or review Gauntlet cards with clear outcomes, a defined persona, and acceptance conditions that can be observed after the task.
Why use it?
It prevents story cards from prescribing clicks or implementation steps, leaving the agent to choose how to achieve the result while keeping success objectively checkable.

Skill for Claude Code

Written for Claude Code: installed under .claude/.

This is prime-radiant-inc/gauntlet's own configuration. It tells Claude Code how to work on gauntlet itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything gauntlet configures →

Reuse

Borrowing it

Nothing to install: this file belongs to prime-radiant-inc/gauntlet. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/prime-radiant-inc/gauntlet/main/.claude/skills/writing-gauntlet-stories/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/prime-radiant-inc/gauntlet

Made for: Claude Code.

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 writing-gauntlet-stories

README.md
[![agentmods](https://agentmods.dev/badge/skills/prime-radiant-inc/gauntlet/writing-gauntlet-stories/github.svg)](https://agentmods.dev/skills/prime-radiant-inc/gauntlet/writing-gauntlet-stories)
Your own site
<a href="https://agentmods.dev/skills/prime-radiant-inc/gauntlet/writing-gauntlet-stories"><img src="https://agentmods.dev/badge/skills/prime-radiant-inc/gauntlet/writing-gauntlet-stories/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for writing-gauntlet-stories

Your own site · 80×15
<a href="https://agentmods.dev/skills/prime-radiant-inc/gauntlet/writing-gauntlet-stories"><img src="https://agentmods.dev/badge/skills/prime-radiant-inc/gauntlet/writing-gauntlet-stories.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 70 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,649 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector pass 7 Sept 2026
How audits are shown
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 5.1 $0.00070 $0.02649
Opus 5 $0.00035 $0.01324
Sonnet 5 $0.00014 $0.00530
Haiku 4.5 $0.00007 $0.00265

Measured 12d ago against content hash 4edfe79031cc, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

writing-gauntlet-stories 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 12d 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.

.claude/skills/writing-gauntlet-stories/SKILL.md · 217 lines

How it starts

The opening of the file, as written. The whole thing — 217 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Writing Gauntlet Stories

Thesis

A Gauntlet story is an outcome, scoped by a persona, graded by observable conditions — with the mechanics deliberately left out.

The agent figures out the how. The card specifies what counts as success in a way that distinguishes a real pass from a confident-sounding bluff.

If you remember nothing else: outcome + persona + observable post-conditions. No clicks, no selectors, no file paths.

Why this differs from the frames you already know

Three familiar frames look right and are wrong in contagious ways. Read these before you start writing — if you skip this, you will write the wrong shape.

  • Agile user story (As a X, I want Y, so that Z). Not this. The persona is not a stakeholder whose needs are being designed for. It is an identity the agent inhabits during the run, used to anchor inference into the context tree.
  • QA step-script (Step 1: navigate. Step 2: type. Step 3: assert). Not this. The whole point of an LLM in the loop is that the agent figures out the path. Enumerating steps recreates exactly what we are trying to escape.
  • BDD Given/When/Then. Not this. Acceptance criteria here are end-state conditions, not consequents of a prescribed sequence. A Gauntlet AC presupposes nothing about how you got there.

If your card reads like any of these, rewrite it.

The rules

Each rule has a rationale (why) and a smell (what it looks like when you are violating it).

1. Open with persona binding, not a profile reference

Every card starts You are <Name>. Never Use the profile at profiles/x/profile.md.

  • Why: the agent reads the context tree and infers which profile to use from the name. That inference is part of the test, especially when decoy profiles exist. Naming the path destroys the inference.
  • Smell: any file path, any reference to a specific profile location, any phrase like "load the profile for…".

2. Describe outcomes, not actions

Verbs sit at intent level (sign in, post, find, confirm) — not mechanics level (click #login, press down-arrow twice, type email then password).

Read the full file on GitHub · 217 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. 12d ago First seen · 217 lines · 70 tokens per session scan A 4edfe79031cc

Subscribe to this mod's changes

writing-gauntlet-stories is a skill published in the GitHub repository prime-radiant-inc/gauntlet (55 stars, last pushed yesterday), licensed Apache-2.0. It adds 70 tokens to every session and 2,649 once invoked, about $0.0003 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.

Related

Other skills, from other repositories

system-prompts

Write system prompts, tool docs, and agent definitions. Project tag conventions + RFC 2119 keywords + dense compression. Use when authoring or editing any prompt the model reads.

can1357/oh-my-pi · 40 tokens

tool-prompt-optimization

Optimize the description prompts an AI agent reads to learn its built-in tools (the .md files under prompts/tools/). Two halves: (1) measure how much of a prompt is already inferable from the tool's JSON parameter schema + name, to prune redundancy with evidence; (2) house authoring rules for what belongs in a tool…

can1357/oh-my-pi · 0 tokens

semantic-compression

Re-encode verbose prose into a dense telegraphic register — punctuation as connectives, label frames, verbless assertions — without losing normativity or precision. Use when compressing system prompts, tool/function descriptions, skill bodies, or agent instructions; reducing token count or context bloat; making…

can1357/oh-my-pi · 77 tokens

effect

Work with Effect v4 / effect-smol TypeScript code in this repo.

bobbyjohnstx/tinycode · 17 tokens

open-source

Documentation reference for writing Python code using the browser-use open-source library. Use this skill whenever the user needs help with Agent, Browser, or Tools configuration, is writing code that imports from browseruse, asks about @sandbox deployment, supported LLM models, Actor API, custom tools, lifecycle…

browser-use/browser-use · 137 tokens

imaging-data-commons

Query and download public cancer imaging data from NCI Imaging Data Commons using idc-index. Use for accessing large-scale radiology (CT, MR, PET) and pathology datasets for AI training or research. No authentication required. Query by metadata, visualize in browser, check licenses.

synthetic-sciences/openscience · 62 tokens