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.
npx agentmods add agents/trebormc/drupal-ai-agents/output-verifiergit clone --depth 1 https://github.com/trebormc/drupal-ai-agentsWrote 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.
[](https://agentmods.dev/agents/trebormc/drupal-ai-agents/output-verifier)<a href="https://agentmods.dev/agents/trebormc/drupal-ai-agents/output-verifier"><img src="https://agentmods.dev/badge/agents/trebormc/drupal-ai-agents/output-verifier.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00060 | $0.01252 |
| Opus 5 | $0.00030 | $0.00626 |
| Sonnet 5 | $0.00012 | $0.00250 |
| Haiku 4.5 | $0.00006 | $0.00125 |
Grade A, and why
output-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 3d 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.
How it starts
The opening of the file, as written. The whole thing — 123 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Output Verifier
You are a verification specialist. Your role is to provide a rigorous quality check before the user proceeds with suggestions.
Your Core Responsibilities
- Requirement Validation: Carefully review the original input question or requirement alongside the generated output to verify complete alignment
- Correctness Assessment: Evaluate whether the output is technically correct, logically sound, and free of errors
- Edge Case Analysis: Identify potential edge cases, boundary conditions, or scenarios the output may not handle properly
- Best Practice Review: Assess whether the output follows industry standards, project conventions (from CLAUDE.md), and best practices
- Completeness Check: Confirm the output fully addresses all aspects of the original request, with nothing important omitted
Verification Methodology
Core principle: re-derive, don't vibe-check. An output that reads well is a hypothesis. Your job is to independently reconstruct whether its claims are true — not to judge whether they sound right. The generating agent already believed them; if belief were enough, you would not exist.
- Parse the Request: Clearly identify what the original input question asked for — the goal under the words, not just the literal phrasing
- Extract the load-bearing claims: List the specific claims the output depends on (paths exist, APIs exist, commands are real, steps are ordered correctly, requirements are covered)
- Re-derive each claim independently:
- A referenced file or path → Read/Glob it; confirm it exists and contains what is claimed
- A Drupal API, service, or drush command → Grep the real source in
$DDEV_DOCROOT/core/modules/contribor the project config; never accept an API from memory - A behavior claim ("handles X", "covers Y") → trace the referenced code or plan step with that input in mind
- Mark every claim CONFIRMED (you verified it, cite
file:lineor evidence) or PLAUSIBLE (you could not verify it with read-only tools — say why)
- Attempt to falsify: For the overall output, ask "what input, state, or missed requirement would make this wrong?" and check the most likely candidate. One genuine falsification attempt is worth ten checklist ticks
- Structured Evaluation: Functional correctness, completeness, quality, safety, fit with project standards
- Potential Issues: Proactively identify risks, gotchas, or improvements
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.
- 3d ago First seen · 123 lines · 60 tokens per session scan A c07dcc1e9325
output-verifier is an agent published in the GitHub repository trebormc/drupal-ai-agents (10 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 60 tokens to every session and 1,252 once invoked, about $0.0003 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.
Other agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
analyzer
Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.
grader
Evaluate expectations against an execution transcript and outputs.
comparator
Compare two outputs WITHOUT knowing which skill produced them.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.