reference

reference is a skill for Claude Code from CrumbtrailDev/crumbtrail-cli. It costs 61 tokens per session (1,392 once invoked), scanned A, original, MIT.

A worked example and template for writing Crumbtrail failure-diagnosis skills. It shows the required sections and explains what evidence a recorded session can or cannot provide.

In plain words
What is it for?
It helps create new failure-archetype skills, describe symptoms, document available evidence, and satisfy the project's skill validation checks.
Why use it?
It gives authors a consistent starting structure and prevents diagnosis guides from claiming causes or signals that the system cannot actually observe.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin.

Part of the crumbtrail-skills plugin — 13 skills shipped together

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 skills/crumbtraildev/crumbtrail-cli/reference
Any agent
npx skills add CrumbtrailDev/crumbtrail-cli --skill reference
Clone the repo
git clone --depth 1 https://github.com/CrumbtrailDev/crumbtrail-cli

Made for: Claude Code.

Or install crumbtrail-skills, the plugin that ships this one along with the rest of its 13 skills.

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 reference

README.md
[![agentmods](https://agentmods.dev/badge/skills/crumbtraildev/crumbtrail-cli/reference.svg)](https://agentmods.dev/skills/crumbtraildev/crumbtrail-cli/reference)
Your own site
<a href="https://agentmods.dev/skills/crumbtraildev/crumbtrail-cli/reference"><img src="https://agentmods.dev/badge/skills/crumbtraildev/crumbtrail-cli/reference.svg" alt="Measured on agentmods" height="20"></a>
Per session 61 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,392 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 5.1 $0.00061 $0.01392
Opus 5 $0.00030 $0.00696
Sonnet 5 $0.00012 $0.00278
Haiku 4.5 $0.00006 $0.00139

Measured yesterday against content hash 6afd76355566, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

reference 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 yesterday.

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.

plugins/crumbtrail-skills/REFERENCE-SKILL.md · 98 lines

How it starts

The opening of the file, as written. The whole thing — 98 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Reference shape for a Crumbtrail archetype skill

Every archetype skill in this plugin uses the five sections below, in this order. The gate at plugins/__tests__/skills.test.ts checks that they are all present, and checks every tool name and parameter name against the live MCP tool table. Copy this file, keep the headings, replace the content.

The worked example running through it is deliberately plain: a user reports that an action in the app failed, and nobody knows yet which archetype it is. That is the state every real skill starts from.

Symptom

State what the reporter or the failing test actually saw, in their words rather than in a diagnosis. One or two sentences.

For the worked example: a user completed an action, the page reported success, and the effect never appeared.

Do not write a cause here. The point of an archetype skill is that several causes produce the same symptom, and the call sequence is what separates them.

What Crumbtrail can see

Name the signals a recorded session actually carries for this archetype, and name the ones it does not. Being explicit about the second list is the whole value of the section. A skill that implies a signal exists sends the agent to run a query that answers nothing, which costs a turn and teaches it a wrong picture of the product.

For the worked example, a session carries console errors and unhandled rejections, every request and response with status and timing, navigation, storage and cookie changes, backend spans where the backend SDK is installed, and database row changes where the database adapter is installed.

It does not carry a stack trace for code that never threw, and it does not carry the value of a variable that was never logged or sent. When an archetype turns on something in that second category, say so in this section and say what the closest available proxy is.

Call sequence

Give the calls as json fenced blocks holding either one object or an array of objects, each shaped {"tool": ..., "params": {...}}. Use real parameter names. The gate rejects a parameter the tool does not accept and rejects a call missing a required parameter, so a sequence that passes the gate is a sequence that will at least dispatch.

Read the full file on GitHub · 98 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. yesterday First seen · 98 lines · 61 tokens per session scan A 6afd76355566

Subscribe to this mod's changes

reference is a skill published in the GitHub repository CrumbtrailDev/crumbtrail-cli (0 stars, last pushed today), licensed MIT. It adds 61 tokens to every session and 1,392 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 2026-09-04.

Related

Other skills, from other repositories

codegraph

Analyze indexed codebases via graph database (neug) and vector index (zvec). Covers call graphs, dependencies, dead code, hotspots, module coupling, architecture reports, semantic search, impact analysis, bug root cause from GitHub issues, class diagrams (UML), and PR review (risk scoring, conflict detection…

QwenLM/qwen-code · 0 tokens

stuck

Diagnose frozen, stuck, or slow Qwen Code sessions on this machine. Scans for problematic processes, high CPU/memory usage, hung subprocesses, and debug logs. Use /stuck or /stuck to focus on a specific process.

QwenLM/qwen-code · 57 tokens

structured-debugging

Hypothesis-driven debugging methodology for hard bugs. Use this skill whenever you're investigating non-trivial bugs, unexpected behavior, flaky tests, or tracing issues through complex systems. Activate proactively when debugging requires more than a quick glance — especially when the first attempt at a fix didn't…

QwenLM/qwen-code · 85 tokens

memory-leak-debug

Diagnose memory leaks in the Qwen Code CLI using heap snapshots and the chrome-devtools CLI. Use when investigating high memory usage, unbounded growth, or suspected object retention issues.

QwenLM/qwen-code · 42 tokens

bugfix

Fix a bug from a GitHub issue, following the reproduce-first workflow. Use when the user asks to fix a bug, investigate a GitHub issue, or debug a user-reported problem. Takes a GitHub issue URL or number as input.

QwenLM/qwen-code · 53 tokens

rust-check

Run cargo check on the current Rust project to find compile errors.

Hmbown/CodeWhale · 15 tokens