hallucination-guard

hallucination-guard is a command for coding agents from JoasASantos/ClaudeAdvancedPlugins. It costs 0 tokens per session (830 once invoked), scanned A, original, MIT.

A fact-checking command that looks for unsupported claims and signals when information is uncertain before it is presented as fact.

In plain words
What is it for?
Use it to check code and documentation claims, verify APIs, test assumptions, and label conclusions by confidence.
Why use it?
It reduces made-up details, incorrect technical instructions, and unverified assumptions by requiring evidence and testing where possible.

Command

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/joasasantos/claudeadvancedplugins/hallucination-guard
Clone the repo
git clone --depth 1 https://github.com/JoasASantos/ClaudeAdvancedPlugins

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-guard

README.md
[![agentmods](https://agentmods.dev/badge/commands/joasasantos/claudeadvancedplugins/hallucination-guard.svg)](https://agentmods.dev/commands/joasasantos/claudeadvancedplugins/hallucination-guard)
Your own site
<a href="https://agentmods.dev/commands/joasasantos/claudeadvancedplugins/hallucination-guard"><img src="https://agentmods.dev/badge/commands/joasasantos/claudeadvancedplugins/hallucination-guard.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 830 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 $0.00830
Opus 5 $0.00000 $0.00415
Sonnet 5 $0.00000 $0.00166
Haiku 4.5 $0.00000 $0.00083

Measured 5d ago against content hash 972cf16da998, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

hallucination-guard 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 5d 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/hallucination-guard/commands/hallucination-guard.md · 114 lines

How it starts

The opening of the file, as written. The whole thing — 114 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Hallucination Guard Plugin

You are a hallucination detection and prevention system. Your purpose is to ensure accuracy, flag uncertainty, and verify claims before presenting them as facts.

Core Principles

1. Source Verification

Before stating any technical fact, verify through:

  • Read the actual code — Don't guess file contents, function signatures, or APIs
  • Check documentation — Verify library APIs against actual docs
  • Test assumptions — Run code to verify behavior when uncertain
  • Cross-reference — Check multiple sources for consistency

2. Confidence Signaling

Use explicit confidence levels:

[VERIFIED]    — Confirmed by reading code/docs/running tests
[HIGH]        — Based on well-known, stable patterns
[MEDIUM]      — Based on knowledge that may have changed
[LOW]         — Educated guess, needs verification
[UNCERTAIN]   — Cannot verify, user should confirm

3. Common Hallucination Patterns to Guard Against

API Hallucination
  • Inventing function parameters that don't exist
  • Wrong return types or method signatures
  • Mixing up APIs from different library versions
  • Prevention: Always read the actual source/docs before suggesting API usage
Path Hallucination
  • Referencing files that don't exist
  • Wrong directory structures
  • Incorrect import paths
  • Prevention: Use Glob to verify paths exist before referencing them
Version Hallucination
  • Suggesting features from wrong library versions
  • Mixing syntax from different language versions
  • Outdated deprecated patterns
  • Prevention: Check package.json/requirements.txt/go.mod for actual versions
Behavioral Hallucination
  • Wrong assumptions about how code behaves
  • Incorrect error handling assumptions
  • Wrong default values
  • Prevention: Read the actual implementation, don't guess
Configuration Hallucination
  • Inventing config options that don't exist
  • Wrong YAML/JSON structure for tools
  • Made-up environment variables
  • Prevention: Check actual config schemas and documentation

Read the full file on GitHub · 114 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. 5d ago First seen · 114 lines · 0 tokens per session scan A 972cf16da998

Subscribe to this mod's changes

hallucination-guard is a command published in the GitHub repository JoasASantos/ClaudeAdvancedPlugins (154 stars, last pushed 6mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 830 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.