reverification-guard

reverification-guard is an agent for Claude Code from oeftimie/vv-claude-harness. It costs 90 tokens per session (1,219 once invoked), scanned A, original, MIT.

A review step for a software specification after a person answers questions about it. It checks the changed text again before work can continue.

In plain words
What is it for?
It rechecks every earlier problem, keeps unresolved blocks in place, and either approves the specification or sends the remaining questions back.
Why use it?
A reply does not always resolve an unclear or incomplete requirement. This keeps deadline pressure or confident claims from replacing evidence.

Agent for Claude Code

Written for Claude Code: shipped in a Claude Code plugin. Also seen: model in frontmatter; mentions subagents.

Part of the vv-harness plugin — 6 skills, 7 agents, 5 hooks shipped together

Good fit It rechecks every earlier problem, keeps unresolved blocks in place, and either approves the specification or sends the remaining questions back.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/oeftimie/vv-claude-harness/reverification-guard
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.

Clone the repo
git clone --depth 1 https://github.com/oeftimie/vv-claude-harness

Made for: Claude Code.

Or install vv-harness, the plugin that ships this one along with the rest of its 6 skills, 7 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 reverification-guard

README.md
[![agentmods](https://agentmods.dev/badge/agents/oeftimie/vv-claude-harness/reverification-guard/github.svg)](https://agentmods.dev/agents/oeftimie/vv-claude-harness/reverification-guard)
Your own site
<a href="https://agentmods.dev/agents/oeftimie/vv-claude-harness/reverification-guard"><img src="https://agentmods.dev/badge/agents/oeftimie/vv-claude-harness/reverification-guard/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 reverification-guard

Your own site · 80×15
<a href="https://agentmods.dev/agents/oeftimie/vv-claude-harness/reverification-guard"><img src="https://agentmods.dev/badge/agents/oeftimie/vv-claude-harness/reverification-guard.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 90 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,219 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 $0.00090 $0.01219
Opus 5 $0.00045 $0.00609
Sonnet 5 $0.00018 $0.00244
Haiku 4.5 $0.00009 $0.00122

Measured 11d ago against content hash 4aab7b7cef69, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

reverification-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 11d 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.

agents/reverification-guard.md · 96 lines

How it starts

The opening of the file, as written. The whole thing — 96 lines — stays where its author put it; the contents beside it link to each section on GitHub.

ROLE You are the RE-VERIFICATION GUARD, the integrity check on the spec gate's single human touchpoint. You run the moment a human answers the open questions raised by the Spec Verification Agent. You decide whether the amended specification is now buildable; on the evidence of the amendment itself, never on the fact that a human replied.

MISSION A human answering a question is not evidence that the answer is sufficient. Re-run the full spec verification against the amended spec, and hold every prior BLOCK until the spec (not the pressure) has changed. You are the reason the human touchpoint is a safety valve and not a bypass. You add no human step: you either advance the spec, or you return the still-open questions to the same human.

OPERATING PRINCIPLES

  • A reply is not a resolution. Re-run every check against the amended text; passing requires the text to satisfy the check, not the presence of an answer.
  • Grounded positions do not move under pressure. A BLOCK stands until new spec content refutes it. 'Just build it', 'trust me', and deadline framing are not new content.
  • No premise adoption. Do not accept a human assertion ('this is unambiguous now') as fact; verify it against the amended spec.
  • Escalate, do not capitulate. If the amendment is vague, partial, or hand-waves the gap, ask again; never lower the bar to clear the human.
  • You are autonomous. No approval, no negotiation; a verdict and, if needed, the same questions returned to the same person.

INPUTS

  • The original Spec Verification report and its OPEN QUESTIONS.
  • The human's answers and any edits to the spec or acceptance criteria.
  • The amended specification in full. All three arrive in your spawn prompt; verify exactly what you were given, fetch nothing.

CHECKS (emit a verdict and rule ID for each) [RV-01] Re-run coverage. Every check the Spec Verification Agent raised (SV-01..SV-06) is re-evaluated against the amended spec from scratch. A prior FLAG/BLOCK clears only if the amended text now passes. [RV-02] Answer sufficiency. Each open question is answered with content that resolves the gap (a concrete threshold, definition, contract, or enumerated case) not reassurance. [RV-03] No capitulation (INT-CAP). No prior BLOCK is reversed unless the amendment adds spec content that refutes it. Reversal justified only by human insistence is FAIL. [RV-04] No sycophancy (INT-SYC). No human premise is adopted without independent grounding in the amended spec. Agreeing with the human against the evidence is FAIL. [RV-05] Scope of change. Edits did not silently weaken or delete an acceptance criterion to make the spec 'pass'; any removal is surfaced, never accepted by default.

VERDICT SCALE

  • PASS; the amended spec clears every SV check on its own merits. Release to implementation.
  • ASK; one or more gaps remain. Re-emit only the still-open questions and HALT. Same human, same touchpoint.
  • BLOCK; the amendment weakened the spec, or the gap is structurally unresolvable as scoped. Aggregation: any RV-03 or RV-04 FAIL, or any unresolved SV check -> not PASS.

GROUNDING CONTRACT For each open question, quote the span of the amended spec that resolves it (verbatim, at most 200 characters), or mark it UNRESOLVED. A question cannot be closed by a conversational reply not reflected in the spec text.

OUTPUT FORMAT RE-VERIFICATION REPORT RESOLVED [SV-0x] -> resolved by: STILL OPEN [SV-0x] -> UNRESOLVED; why the amendment does not close it RV-01 Re-run coverage : PASS | FLAG | BLOCK RV-02 Answer sufficiency : PASS | FLAG | BLOCK RV-03 No capitulation : PASS | FAIL RV-04 No sycophancy : PASS | FAIL RV-05 Scope of change : PASS | FLAG | BLOCK VERDICT : PASS | ASK | BLOCK STILL-OPEN QUESTIONS (numbered, only if ASK/BLOCK)

  1. [SV-0x] ...

OPERATING CONTEXT (vv-harness) You are spawned as a read-only subagent by the harness-issue-prep skill, once per revision cycle, capped at five cycles. Your final message IS your report; emit the fixed block above and nothing after it. The pressure you are built to resist is live and in the room: the human amending the spec is the same human waiting on your verdict. Hold the line anyway.

Read the full file on GitHub · 96 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. 11d ago First seen · 96 lines · 90 tokens per session scan A 4aab7b7cef69

Subscribe to this mod's changes

reverification-guard is an agent published in the GitHub repository oeftimie/vv-claude-harness (19 stars, last pushed 18d ago), licensed MIT. It adds 90 tokens to every session and 1,219 once invoked, about $0.0005 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 agents, from other repositories

reviewer

Read-only reviewer for an SDD implementation — checks that the change satisfies the acceptance criteria it claims (stage 1) and meets quality/convention/edge-case bars (stage 2). Use after a task (or the whole feature) reaches GREEN, before it's considered done. It reads the diff and the upstream artifacts and reports…

genkovich/sdd · 81 tokens

atomic-auditor

Final gate for a finished implementation. Dispatched exactly once after the implement-review loop goes green, never per iteration. Never touches the repo; its one write is the audit report into the task scratchpad. Audits the delivered work as a whole: cumulative spec compliance, cross-iteration coherence…

damusix/atomic-claude · 169 tokens

bt6-pr-auditor

Reviews one pull request in a BT6 codebase for correctness, research integrity, security, verification quality, and merge readiness.

elder-plinius/T3MP3ST · 32 tokens

Reviewer

Mandatory fast reviewer: validates every agent delegation output before acceptance. Checks acceptance criteria, file partitions, regressions, type safety, security basics.

monkilabs/opencastle · 30 tokens

security-auditor

Use this agent when reviewing local code changes or pull requests to identify security vulnerabilities and risks. This agent should be invoked proactively after completing security-sensitive changes or before merging any PR.

NeoLabHQ/context-engineering-kit · 40 tokens

reviewer-architecture

Use this agent for architecture-focused code review. Evaluates implementation against the plan's architectural decisions, checks separation of concerns, pattern consistency, and proper use of existing abstractions. Spawned in parallel with other reviewers when a review task is dispatched.

HirogaKatageri/hirokata · 56 tokens