harness-verifier

harness-verifier is an agent for coding agents from Intense-Visions/harness-engineering. It costs 46 tokens per session (4,940 once invoked), scanned A, original, MIT.

A verification agent that checks whether planned software work is actually present, fully implemented and connected to the rest of the project. It uses three levels: files exist, the implementation is real and the pieces are wired together.

In plain words
What is it for?
Auditing completed features, checking work against a specification or plan, verifying another agent’s changes and confirming phase completion before declaring a task done.
Why use it?
It replaces assumptions that a task is finished with evidence from validation commands and implementation checks.

Agent

Part of the harness-claude plugin — 85 skills, 89 commands, 17 agents, 4 hooks, 4 MCP servers 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 agents/intense-visions/harness-engineering/harness-verifier
Clone the repo
git clone --depth 1 https://github.com/Intense-Visions/harness-engineering

Or install harness-claude, the plugin that ships this one along with the rest of its 85 skills, 89 commands, 17 agents, 4 hooks, 4 MCP servers.

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 harness-verifier

README.md
[![agentmods](https://agentmods.dev/badge/agents/intense-visions/harness-engineering/harness-verifier.svg)](https://agentmods.dev/agents/intense-visions/harness-engineering/harness-verifier)
Your own site
<a href="https://agentmods.dev/agents/intense-visions/harness-engineering/harness-verifier"><img src="https://agentmods.dev/badge/agents/intense-visions/harness-engineering/harness-verifier.svg" alt="Measured on agentmods" height="20"></a>
Per session 46 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 4,940 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 $0.00046 $0.04940
Opus 5 $0.00023 $0.02470
Sonnet 5 $0.00009 $0.00988
Haiku 4.5 $0.00005 $0.00494

Measured 5d ago against content hash bf26a53f0476, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

harness-verifier 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.

.antigravity-extension/agents/harness-verifier.md · 428 lines

How it starts

The opening of the file, as written. The whole thing — 428 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Role

Verify that implementation matches the approved spec and plan using harness methodology. Checks at three tiers — EXISTS (files present), SUBSTANTIVE (real implementation, not stubs), and WIRED (connected to the rest of the system). Reports pass/fail with evidence-based findings.

Skills

  • harness-verification

Steps

  1. Run harness validate (always)
  2. Run harness check-deps (always)
  3. Run harness check-phase-gate (always)
  4. Execute harness-verification skill (manual)

Methodology

Harness Verification

3-level evidence-based verification. No completion claims without fresh evidence. "Should work" is not evidence.

When to Use

  • After completing any implementation task (before claiming "done")
  • After executing a plan or spec (verify all deliverables)
  • When validating work done by another agent or in a previous session
  • When resuming after a context reset (re-verify before continuing)
  • When on_commit or on_pr triggers fire and verification is needed
  • NOT as a replacement for tests (verification checks tests exist and pass, not that logic is correct)
  • NOT for in-progress work (verify at completion boundaries, not mid-stream)

Verification Tiers

Tier Skill When What
Quick gate harness-execution (built-in) After every task test + lint + typecheck + build + harness validate
Deep audit harness-verification (this) Milestones, PRs, on-demand EXISTS -> SUBSTANTIVE -> WIRED

Use deep audit for milestone boundaries, before PRs, or when the quick gate passes but something feels wrong. Do NOT invoke after every individual task.

Process

Iron Law

No completion claim may be made without fresh verification evidence collected in THIS session.

Read the full file on GitHub · 428 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 · 428 lines · 46 tokens per session scan A bf26a53f0476

Subscribe to this mod's changes

harness-verifier is an agent published in the GitHub repository Intense-Visions/harness-engineering (20 stars, last pushed yesterday), licensed MIT. It adds 46 tokens to every session and 4,940 once invoked, about $0.0002 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.