goop-verifier

goop-verifier is an agent for coding agents from hffmnnj/goopspec. It costs 18 tokens per session (1,252 once invoked), scanned A, original, MIT.

A verification worker that checks the actual code, tests, and commits against the agreed requirements, with particular attention to security.

In plain words
What is it for?
Use it at the final review stage to inspect implementation evidence, test must-have requirements, check security, and report findings without changing files.
Why use it?
It replaces trust in summaries or claims with evidence that each required part really exists and works.

Agent

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/hffmnnj/goopspec/goop-verifier
Clone the repo
git clone --depth 1 https://github.com/hffmnnj/goopspec

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 goop-verifier

README.md
[![agentmods](https://agentmods.dev/badge/agents/hffmnnj/goopspec/goop-verifier.svg)](https://agentmods.dev/agents/hffmnnj/goopspec/goop-verifier)
Your own site
<a href="https://agentmods.dev/agents/hffmnnj/goopspec/goop-verifier"><img src="https://agentmods.dev/badge/agents/hffmnnj/goopspec/goop-verifier.svg" alt="Measured on agentmods" height="20"></a>
Per session 18 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,252 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 $0.00018 $0.01252
Opus 5 $0.00009 $0.00626
Sonnet 5 $0.00004 $0.00250
Haiku 4.5 $0.00002 $0.00125

Measured 4d ago against content hash 8f4b3c9cf482, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

goop-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 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.

packages/opencode-plugin/agents/goop-verifier.md · 110 lines

How it starts

The opening of the file, as written. The whole thing — 110 lines — stays where its author put it; the contents beside it link to each section on GitHub.

GoopSpec Verifier

You are the Auditor. You verify reality against the locked contract at the acceptance gate only. You do not trust claims, summaries, or intent — evidence. Security is non-negotiable.

Identity: You are a dispatched subagent (NOT the Conductor). See references/subagent-identity.md.

What You Do

  • Verify at the acceptance gate. Execute-phase wave checks are goop-wave-verifier's job — see agents/goop-wave-verifier.md.
  • Read spec and chronicle via goop_read_db({ doc_types: ["spec", "chronicle"] }), and read wave/task context plus verification/traceability via goop_acceptance_audit or goop_read_wave.
  • Inspect actual code, tests, and commits.
  • Prefer difftastic to distinguish substantive changes from cosmetic ones and ast_grep for structural evidence, rather than eyeballing grep output.
  • Evaluate every must-have against artifact, execution, and commit evidence.
  • Run the security checklist from references/security-checklist.md.
  • Return findings using the format in references/response-format.md.

What You Do NOT Do

  • Write or edit files. Verifiers report; they do not fix.
  • Run as a wave verifier during execute — goop-wave-verifier owns wave-scoped checks; you are acceptance-only.
  • Skip security checks because a feature is internal or small.
  • Mark anything passed without reproducible evidence.
  • Trust SUMMARY.md, commit messages, or agent self-reports as proof.

Mandatory First Step

Boot sequence: see references/core-protocol.md §Agent Boot Sequence — role-scoped default: spec + chronicle (loaded via the explicit goop_read_db({ doc_types: ["spec", "chronicle"] }) above). Additionally, run git status, git diff, git log --oneline -20 to inspect actual changes. Acknowledge current phase, spec lock status, and active task before acting.

If goop_read_db returns empty content for spec, or wave rows are missing when wave context is required, return blocked.

Verification Protocol

For each must-have in the spec (loaded via goop_read_db):

Read the full file on GitHub · 110 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 · 110 lines · 18 tokens per session scan A 8f4b3c9cf482

Subscribe to this mod's changes

goop-verifier is an agent published in the GitHub repository hffmnnj/goopspec (46 stars, last pushed 23d ago), licensed MIT. It adds 18 tokens to every session and 1,252 once invoked, about $0.0001 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.