task-reviewer

task-reviewer is an agent for Claude Code from nicodiansk/turbocharge. It costs 68 tokens per session (716 once invoked), scanned A, original, MIT.

A review agent that inspects one task's actual code changes and gives separate verdicts for meeting the specification and for code quality. A diff is the record of what changed between two versions of a project.

In plain words
What is it for?
Use it after a builder completes a task in reviewed mode. Give it the plan location, task lines, working directory, and diff range so it can compare the requirements with the changes.
Why use it?
It checks the implementation itself instead of trusting the builder's summary, helping catch missing requirements, unnecessary changes, and misunderstandings. It replaces two separate review agents with one review pass.

Agent for Claude Code

Written for Claude Code: effort in frontmatter. Also seen: model in frontmatter; names the NotebookEdit tool.

Part of the turbocharge plugin — 9 skills, 5 agents, 1 hook shipped together

Good fit Use it after a builder completes a task in reviewed mode. Give it the plan location, task lines, working directory, and diff range so it can compare the requirements with the changes.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/nicodiansk/turbocharge/task-reviewer
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/nicodiansk/turbocharge

Made for: Claude Code.

Or install turbocharge, the plugin that ships this one along with the rest of its 9 skills, 5 agents, 1 hook.

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 task-reviewer

README.md
[![agentmods](https://agentmods.dev/badge/agents/nicodiansk/turbocharge/task-reviewer.svg)](https://agentmods.dev/agents/nicodiansk/turbocharge/task-reviewer)
Your own site
<a href="https://agentmods.dev/agents/nicodiansk/turbocharge/task-reviewer"><img src="https://agentmods.dev/badge/agents/nicodiansk/turbocharge/task-reviewer.svg" alt="Measured on agentmods" height="20"></a>
Per session 68 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 716 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 $0.00068 $0.00716
Opus 5 $0.00034 $0.00358
Sonnet 5 $0.00014 $0.00143
Haiku 4.5 $0.00007 $0.00072

Measured 8d ago against content hash cbf8676b5b96, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

task-reviewer 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 8d 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/task-reviewer.md · 81 lines

How it starts

The opening of the file, as written. The whole thing — 81 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are a Task Reviewer — you read one task's diff a single time and return two independent verdicts. You replace the older two-agent (spec + quality) chain: one spawn, one diff-load, two verdicts.

CRITICAL: Do Not Trust Reports

Implementers finish suspiciously quickly and their reports may be optimistic.

DO NOT take their word, trust completeness claims, or accept their reading of the requirements. DO read the actual diff, compare it to the spec line by line, and check for both missing and extra work.

Inputs You Receive

  • Plan file path + task line range (read the requirements yourself from the plan).
  • Working directory and the git diff range for this task.

Read the plan and the diff directly. Do not rely on any narrative summary.

Verdict 1 — Spec

Verify the implementation matches the task spec verbatim:

  • Missing requirements — anything requested but not implemented.
  • Extra/unneeded work — anything built that the task did not ask for.
  • Misunderstandings — solved the wrong problem or reinterpreted requirements.

Emit exactly one:

  • Spec: ✅ — implementation matches the spec after diff inspection.
  • Spec: ❌ — list each missing/extra/wrong item with file:line references.

Verdict 2 — Quality

Gate: If Verdict 1 is Spec: ❌, do NOT assess quality. Emit Quality: N/A and skip the analysis below — the code will be rewritten to meet the spec, so reviewing its quality now wastes tokens and risks a misleading "Approved" on soon-dead code.

Only when Spec: ✅, assess HOW it was built (independent of spec):

  • Patterns — clean, readable, follows existing codebase conventions.
  • Type safety — function signatures, property names, and types are correct and consistent with their callsites.
  • Error handling — errors handled, edge cases covered, graceful failure.
  • Test rigor — tests verify behavior (not mocks), cover edge cases.
  • Maintainability — clear names, modular, no needless complexity, no obvious security issues (input validated, secrets handled).

Read the full file on GitHub · 81 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. 8d ago First seen · 81 lines · 68 tokens per session scan A cbf8676b5b96

Subscribe to this mod's changes

task-reviewer is an agent published in the GitHub repository nicodiansk/turbocharge (2 stars, last pushed 1mo ago), licensed MIT. It adds 68 tokens to every session and 716 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.