anti-hallucination-gates

anti-hallucination-gates is a skill for Claude Code from Redtropig/harness-anchor. It costs 72 tokens per session (2,530 once invoked), scanned A, original, MIT.

A checklist for verifying claims that work is finished, fixed, passing, or unavailable. It requires concrete evidence for positive claims and a stated search scope and date for claims that something is absent.

In plain words
What is it for?
Use it before reporting a fix, completed task, passing build or tests, missing capability, unavailable tool, or failed search.
Why use it?
It reduces premature conclusions by making agents support their statements with test results, logs, files, or clearly described searches.

Skill for Claude Code

Written for Claude Code: ${CLAUDE_PLUGIN_ROOT} variable. Also seen: mentions CLAUDE.md; mentions subagents; mentions AGENTS.md.

Runs only inside its plugin — its command needs a path that Claude Code sets for a plugin’s own hooks and for nothing else. Install the plugin, not this.

Part of the harness-anchor plugin — 14 skills, 9 commands, 5 agents, 5 hooks shipped together

Good fit Use it before reporting a fix, completed task, passing build or tests, missing capability, unavailable tool, or failed search.

Compare 6 skills from other repositories ↓
Install

Getting it into your agent

This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.

Claude Code
/plugin marketplace add Redtropig/harness-anchor
Claude Code
/plugin install harness-anchor

Made for: Claude Code.

Or install harness-anchor, the plugin that ships this one along with the rest of its 14 skills, 9 commands, 5 agents, 5 hooks.

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 anti-hallucination-gates

README.md
[![agentmods](https://agentmods.dev/badge/skills/redtropig/harness-anchor/anti-hallucination-gates/github.svg)](https://agentmods.dev/skills/redtropig/harness-anchor/anti-hallucination-gates)
Your own site
<a href="https://agentmods.dev/skills/redtropig/harness-anchor/anti-hallucination-gates"><img src="https://agentmods.dev/badge/skills/redtropig/harness-anchor/anti-hallucination-gates/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 anti-hallucination-gates

Your own site · 80×15
<a href="https://agentmods.dev/skills/redtropig/harness-anchor/anti-hallucination-gates"><img src="https://agentmods.dev/badge/skills/redtropig/harness-anchor/anti-hallucination-gates.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 72 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,530 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.
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 $0.00072 $0.02530
Opus 5 $0.00036 $0.01265
Sonnet 5 $0.00014 $0.00506
Haiku 4.5 $0.00007 $0.00253

Measured 12d ago against content hash 68cda5de3133, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

anti-hallucination-gates 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.

skills/anti-hallucination-gates/SKILL.md · 188 lines

How it starts

The opening of the file, as written. The whole thing — 188 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Anti-Hallucination Gates

The most common agent failure is declaring victory too early. This skill is the gate between "I think it works" and "evidence proves it works".

The Iron Rule

No claim of "done" / "fixed" / "passing" without a concrete artifact path proving it.

A claim like "I fixed the bug" is incomplete. The proper form:

"Fixed by commit abc123. Test tests/foo.test.ts::handles-edge-case now passes (output captured in .harness-anchor/test-2026-05-28.log). Build still passes (.build/last-build.log exit 0)."

The Iron Rule runs both ways

The rule above governs claims that something IS. Claims that something IS NOT are the same failure with the sign flipped, and until v0.17.0 this skill said nothing about them.

No claim that something is absent without stating the scope you actually searched and when you looked.

"clang-tidy isn't installed on this machine" is incomplete in exactly the way "I fixed the bug" is incomplete. The proper form:

"searched PATH + the VS-bundled LLVM and CMake directories, not found (as of 2026-07-29)"

Two classes, two probes

Class Sounds like What counts as evidence
Capability "not installed", "not available here", "this platform can't do X" The search chain's coverage plus the date you looked. C/C++ tools: bash ${CLAUDE_PLUGIN_ROOT}/scripts/cpp-tool-discovery.sh <tool> — its NOT_FOUND line already enumerates its own scope. Other ecosystems: command -v plus that platform's known install locations.
Search "there's no such function", "nothing tests this", "that option isn't in the config" The command you actually ran and its coverage: which paths, which glob, case-sensitive or not, whether ignored files were included.

An empty result proves only that your search found nothing. That is a different proposition from the thing is absent, and conflating the two is what this section exists to stop.

The "there is no" detector

Read the full file on GitHub · 188 lines

Files

What ships with it

1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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 · 188 lines · 0 tokens per session scan A 68cda5de3133

Subscribe to this mod's changes

anti-hallucination-gates is a skill published in the GitHub repository Redtropig/harness-anchor (13 stars, last pushed 1mo ago), licensed MIT. It adds 72 tokens to every session and 2,530 once invoked, about $0.0004 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

testing-e2e

End-to-end testing patterns with Playwright — page objects, AI agent testing, visual regression, accessibility testing with axe-core, and CI integration. Use when writing E2E tests, setting up Playwright, implementing visual regression, or testing accessibility.

yonatangross/orchestkit · 55 tokens

testing-llm

LLM and AI testing patterns — mock responses, evaluation with DeepEval/RAGAS, structured output validation, and agentic test patterns (generator, healer, planner). Use when testing AI features, validating LLM outputs, or building evaluation pipelines.

yonatangross/orchestkit · 55 tokens

bare-eval

Run isolated eval and grading calls using CC 2.1.81 --bare mode. Constructs claude -p --bare invocations for skill evaluation, trigger testing, and LLM grading without plugin/hook interference. Use when running eval pipelines, grading skill outputs, benchmarking prompt quality, or testing trigger accuracy in isolation.

yonatangross/orchestkit · 69 tokens

golden-dataset

Golden dataset lifecycle patterns for curation, versioning, quality validation, and CI integration. Use when building evaluation datasets, managing dataset versions, validating quality scores, or integrating golden tests into pipelines.

yonatangross/orchestkit · 44 tokens

paired-probe

Refuse a verdict a probe did not earn. Runs a check where the fault IS present and where it is NOT, and blocks the answer when both arms print the same thing, because a check that cannot disagree with you has measured nothing. Also catches the zero-sample sweep that reads as "clean" and the swallowed error that reads…

yonatangross/orchestkit · 100 tokens

testing-patterns

Redirect — testing-patterns was split into 5 focused sub-skills. Use when looking for testing-patterns, writing tests, or test automation. Redirects to testing-unit, testing-e2e, testing-integration, testing-llm, or testing-perf.

yonatangross/orchestkit · 59 tokens