fuzz

fuzz is a command for coding agents from robzilla1738/roberts-skills. It costs 14 tokens per session (267 once invoked), scanned A, original, MIT.

A command for creating and running generated-input tests, fuzz tests, or comparisons between implementations. It can reduce a failure to a small regression test, but it writes test or harness code rather than product fixes.

In plain words
What is it for?
It is for hardening a named function or surface, checking an invariant, and generating a test harness using the project's existing test tools.
Why use it?
It helps find failures caused by unusual inputs and makes confirmed failures easier to reproduce.

Command

Installs and runs on its own, but its text points at files inside its plugin — anything it tells you to read at a ${CLAUDE_PLUGIN_ROOT} path is only there once the plugin is installed. Installing the plugin gets both.

Part of the bughunt-suite plugin — 3 skills, 4 commands 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 commands/robzilla1738/roberts-skills/fuzz
Clone the repo
git clone --depth 1 https://github.com/robzilla1738/roberts-skills

Or install bughunt-suite, the plugin that ships this one along with the rest of its 3 skills, 4 commands.

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 fuzz

README.md
[![agentmods](https://agentmods.dev/badge/commands/robzilla1738/roberts-skills/fuzz.svg)](https://agentmods.dev/commands/robzilla1738/roberts-skills/fuzz)
Your own site
<a href="https://agentmods.dev/commands/robzilla1738/roberts-skills/fuzz"><img src="https://agentmods.dev/badge/commands/robzilla1738/roberts-skills/fuzz.svg" alt="Measured on agentmods" height="20"></a>
Per session 14 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 267 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 $0.00014 $0.00267
Opus 5 $0.00007 $0.00133
Sonnet 5 $0.00003 $0.00053
Haiku 4.5 $0.00001 $0.00027

Measured 4d ago against content hash 7f79eb7e6b3f, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

fuzz 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.

plugins/bughunt-suite/commands/fuzz.md · 15 lines

What it actually says

Fuzz / property-test: $ARGUMENTS

Load and follow the fuzz skill at ${CLAUDE_PLUGIN_ROOT}/skills/fuzz/SKILL.md:

  • Discover the target — a named function/surface, or run python3 ${CLAUDE_PLUGIN_ROOT}/skills/bughunt/scripts/bughunt.py census --functions to shortlist fuzzable pure functions.
  • Pick the technique (property-based, fuzzing, differential, or metamorphic) and design the oracle (round-trip, invariant, never-crash, agreement) — this is where the power is.
  • Generate a harness in the project's existing test runner/library; ask before installing any new test dependency. Run under sanitizers where available.
  • Execute it. On a failure: reproduce, shrink to the smallest input, keep or present it as a named regression test, and emit a findings entry marked verified.verdict: upheld. With no test infra, generate + present the harness clearly marked not executed.

Writes test/harness code, not product fixes. If $ARGUMENTS is empty, ask which function/surface to harden and what invariant should hold.

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 · 15 lines · 14 tokens per session scan A 7f79eb7e6b3f

Subscribe to this mod's changes

fuzz is a command published in the GitHub repository robzilla1738/roberts-skills (5 stars, last pushed 2mo ago), licensed MIT. It adds 14 tokens to every session and 267 once invoked, about $0.0001 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 2026-08-31.