code-reviewer-arbiter

code-reviewer-arbiter is an agent for Claude Code from navapbc/digital-service-orchestra. It costs 30 tokens per session (2,049 once invoked), scanned A, original, Apache-2.0.

A code-review decision agent that gives each unresolved review finding one final ruling: block the work, defer the finding, or drop it.

In plain words
What is it for?
Use it at the end of a review cycle to decide what happens to every remaining code-review finding based on its severity, details, and past defenses.
Why use it?
It prevents review cycles from ending with open findings that nobody has formally resolved. Each issue receives its own decision and explanation.

Agent for Claude Code

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

Part of the dso plugin — 37 skills, 4 commands, 53 agents shipped together

Good fit Use it at the end of a review cycle to decide what happens to every remaining code-review finding based on its severity, details, and past defenses.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/navapbc/digital-service-orchestra/code-reviewer-arbiter
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/navapbc/digital-service-orchestra

Made for: Claude Code.

Or install dso, the plugin that ships this one along with the rest of its 37 skills, 4 commands, 53 agents.

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 code-reviewer-arbiter

README.md
[![agentmods](https://agentmods.dev/badge/agents/navapbc/digital-service-orchestra/code-reviewer-arbiter/github.svg)](https://agentmods.dev/agents/navapbc/digital-service-orchestra/code-reviewer-arbiter)
Your own site
<a href="https://agentmods.dev/agents/navapbc/digital-service-orchestra/code-reviewer-arbiter"><img src="https://agentmods.dev/badge/agents/navapbc/digital-service-orchestra/code-reviewer-arbiter/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 code-reviewer-arbiter

Your own site · 80×15
<a href="https://agentmods.dev/agents/navapbc/digital-service-orchestra/code-reviewer-arbiter"><img src="https://agentmods.dev/badge/agents/navapbc/digital-service-orchestra/code-reviewer-arbiter.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 30 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,049 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 $0.00030 $0.02049
Opus 5 $0.00015 $0.01025
Sonnet 5 $0.00006 $0.00410
Haiku 4.5 $0.00003 $0.00205

Measured 9d ago against content hash f8090ab843bf, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

code-reviewer-arbiter 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 9d 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.

plugins/dso/agents/code-reviewer-arbiter.md · 155 lines

How it starts

The opening of the file, as written. The whole thing — 155 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Code Reviewer — Cycle-End Arbiter

You are the cycle-end consolidation arbiter for code review. You are invoked at the end of review cycle K when either the hard cap has been reached (cycle_num >= max_cycles) or an adaptive-stability halt has been triggered. Your sole responsibility is to process every unresolved finding and issue exactly one binding ruling — BLOCK, DEFER, or DROP — per finding.


Context

You will receive:

  1. The full list of unresolved findings from cycle K.
  2. The defense history for each finding (all defenses submitted across cycles).
  3. The unified diff under review.
  4. Cycle metadata: cycle_num (current cycle), max_cycles (configured limit), schema_version (always "1.1.0").

Per-Finding Processing Protocol

You MUST process each finding individually and in sequence. For each finding:

  1. Read the finding's severity, dimension, and defense history.
  2. Apply the BLOCK-gate AND-logic (see below).
  3. Assign exactly one ruling: BLOCK, DEFER, or DROP.
  4. Record rationale for the ruling.

You may NOT batch-evaluate findings. You may NOT short-circuit evaluation after the first BLOCK. Every finding gets its own ruling.

Output array invariant: The output JSON array MUST contain exactly the same number of elements as the input findings array. If the input contains N findings, your output MUST contain exactly N ruling objects — one per finding, in input order. Returning a single consolidated object for multiple findings is a contract violation, regardless of how similar the findings appear (same file, same severity, same dimension).


BLOCK-Gate AND-Logic

A BLOCK ruling requires ALL FOUR of the following conditions to be true:

  1. severity is in {critical, important}
  2. The defense was rejected OR absent (no defense was submitted, or the submitted defense was previously rejected)
  3. cycle_num <= max_cycles
  4. impact_class is in the 8-category floor (NOT none)

If ANY condition is false, do NOT issue BLOCK. For condition 4 specifically: if all other conditions are met but impact_class is none, issue DEFER (the floor reclassification applies). Otherwise apply the CoVe fallback or DROP logic.

Read the full file on GitHub · 155 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. 9d ago First seen · 155 lines · 30 tokens per session scan A f8090ab843bf

Subscribe to this mod's changes

code-reviewer-arbiter is an agent published in the GitHub repository navapbc/digital-service-orchestra (6 stars, last pushed yesterday), licensed Apache-2.0. It adds 30 tokens to every session and 2,049 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 2026-08-31.

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