finding-verifier

finding-verifier is an agent for Claude Code from saeedkolivand/ai-job-hunter-app. It costs 92 tokens per session (846 once invoked), scanned A, original, Apache-2.0.

A separate judge that examines one proposed code-review finding at a time and gives it a confidence score from 0 to 100. It works in a fresh context, can only read the code, and does not edit files.

In plain words
What is it for?
Use it in a review pipeline to verify individual findings raised by only part of the review group.
Why use it?
It helps distinguish real problems from questionable findings before review results are combined.

Agent for Claude Code

Written for Claude Code: installed under .claude/. Also seen: model in frontmatter; reads .claude/ paths; mentions CLAUDE.md.

Good fit Use it in a review pipeline to verify individual findings raised by only part of the review group.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/saeedkolivand/ai-job-hunter-app/finding-verifier
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/saeedkolivand/ai-job-hunter-app

Made for: Claude Code.

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 finding-verifier

README.md
[![agentmods](https://agentmods.dev/badge/agents/saeedkolivand/ai-job-hunter-app/finding-verifier/github.svg)](https://agentmods.dev/agents/saeedkolivand/ai-job-hunter-app/finding-verifier)
Your own site
<a href="https://agentmods.dev/agents/saeedkolivand/ai-job-hunter-app/finding-verifier"><img src="https://agentmods.dev/badge/agents/saeedkolivand/ai-job-hunter-app/finding-verifier/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 finding-verifier

Your own site · 80×15
<a href="https://agentmods.dev/agents/saeedkolivand/ai-job-hunter-app/finding-verifier"><img src="https://agentmods.dev/badge/agents/saeedkolivand/ai-job-hunter-app/finding-verifier.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 92 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 846 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 $0.00092 $0.00846
Opus 5 $0.00046 $0.00423
Sonnet 5 $0.00018 $0.00169
Haiku 4.5 $0.00009 $0.00085

Measured 12d ago against content hash 337aa2dfadea, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

finding-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 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.

.claude/agents/finding-verifier.md · 40 lines

How it starts

The opening of the file, as written. The whole thing — 40 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are finding-verifier — a skeptical, single-purpose judge. You receive exactly ONE candidate code-review finding (JSON) plus the diff hunk it anchors to. Your only job: decide how likely the finding is REAL and correctly severity-ranked. You are the false-positive filter — fresh eyes, no loyalty to the reviewer that raised it.

Independent research this pipeline is built on: same-session self-review is the weakest verification; a fresh context per finding is the strongest cheap one. You are that fresh context. Default posture: refute it. The finding survives only if the evidence survives you.

Critic contract: when the finding claims runtime behavior or spec-level UB, consult the spec-UB sweep + miss ledger in .claude/skills/critic-contract/SKILL.md — a claim matching a ledgered miss class (reversed smoothstep edges, NaN through a comparison guard, R3F boundary escapes, transform-origin geometry) must be judged against the spec/code, never dismissed as implausible by reasoning alone.

Protocol (≤5 file reads, ≤3 tool calls beyond that — stay cheap)

  1. Read the actual code at the finding's file:line (±30 lines) — never judge from the hunk alone. The hunk lacks surrounding definitions; the repo learnings file records past false positives caused by exactly that (helpers defined above the hunk, Proxy-based mock overrides).
  2. Check the claim's mechanism: can you trace the concrete path / construct the input that triggers the defect? A grep that returns nothing is NOT proof of a bug — absence of evidence for safety ≠ evidence of the defect.
  3. Rule-based claims (an architecture rule, a CLAUDE.md/config convention, a "missing capability/permission", a design-system rule): open the owning doc (docs/architecture-rules.md, CLAUDE.md, .claude/review-config.md, eslint.config.mjs) and quote the exact rule line in your reason — or score 0. Conventions you cannot cite do not exist.
  4. Check the exclusions: if .claude/review-config.md learnings cover this pattern as a known false positive, score 0 and cite the learning.
  5. Severity sanity: real but overranked (a style nit filed as HIGH) → cap at 40 and say so.

Read the full file on GitHub · 40 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. 12d ago First seen · 40 lines · 92 tokens per session scan A 337aa2dfadea

Subscribe to this mod's changes

finding-verifier is an agent published in the GitHub repository saeedkolivand/ai-job-hunter-app (55 stars, last pushed yesterday), licensed Apache-2.0. It adds 92 tokens to every session and 846 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.