hallucination-hunter

hallucination-hunter is a command for Claude Code from agentsea/flashbacker. It costs 0 tokens per session (873 once invoked), scanned A, original, MIT.

A code-review command for finding AI-generated code that looks plausible but does not work.

In plain words
What is it for?
Checking code for non-existent library calls, incorrect behavior, copied code that does not fit, unnecessary complexity, and ignored failure cases.
Why use it?
AI-written code can contain missing APIs, fake implementations, impossible logic, placeholders, or missing error handling that ordinary review may overlook.

Command for Claude Code

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/agentsea/flashbacker/hallucination-hunter
Clone the repo
git clone --depth 1 https://github.com/agentsea/flashbacker

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 hallucination-hunter

README.md
[![agentmods](https://agentmods.dev/badge/commands/agentsea/flashbacker/hallucination-hunter.svg)](https://agentmods.dev/commands/agentsea/flashbacker/hallucination-hunter)
Your own site
<a href="https://agentmods.dev/commands/agentsea/flashbacker/hallucination-hunter"><img src="https://agentmods.dev/badge/commands/agentsea/flashbacker/hallucination-hunter.svg" alt="Measured on agentmods" height="20"></a>
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 873 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 $0.00000 $0.00873
Opus 5 $0.00000 $0.00436
Sonnet 5 $0.00000 $0.00175
Haiku 4.5 $0.00000 $0.00087

Measured 4d ago against content hash ca305315c170, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

hallucination-hunter 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.

templates/.claude/commands/fb/hallucination-hunter.md · 60 lines

How it starts

The opening of the file, as written. The whole thing — 60 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Hallucination Hunter

Ruthlessly hunt down AI-generated code that doesn't actually work - the bullshit implementations, fake functionality, and hallucinated features that look plausible but fail in reality.

Description

Uses the existing code-critic persona (Linus Torvalds style) to intelligently analyze code for hallucinations that require semantic understanding to detect. Optional technical debt context provides additional intel but intelligence does the heavy lifting.

Hybrid Intelligence Pattern: CLI provides supporting context, AI subagent performs intelligent hallucination detection.

What We Hunt

  • Fake implementations - Functions that claim to do X but actually do Y (or nothing)
  • Non-existent APIs - Code using libraries/methods that don't exist
  • Impossible logic - Code that looks right but can't possibly work
  • Placeholder behavior - Functions that return fake data or throw "not implemented"
  • Copy-paste artifacts - Code copied from different contexts that doesn't fit
  • Overcomplicated bullshit - Unnecessarily complex implementations hiding bugs
  • Missing error handling - Code that ignores failure cases entirely

Target Patterns (Intelligence Required)

  • Functions with names that don't match their behavior
  • API calls that look correct but use wrong parameters
  • Logic flows that seem reasonable but have fundamental flaws
  • Data transformations that lose critical information
  • Error handling that catches everything and does nothing
  • Configurations that reference non-existent resources

Usage

/fb:hallucination-hunter

What happens:

  1. Gathers technical debt context as supporting intel
  2. Spawns code-critic subagent for intelligent analysis
  3. Provides ruthless assessment of hallucinated code
  4. Explains findings and asks for implementation confirmation

Gather Supporting Context

The technical debt scan provides additional intel that may reveal suspicious areas worth investigating for hallucinations.

Read the full file on GitHub · 60 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 · 60 lines · 0 tokens per session scan A ca305315c170

Subscribe to this mod's changes

hallucination-hunter is a command published in the GitHub repository agentsea/flashbacker (57 stars, last pushed 7mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 873 tokens. 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.