review-synthesizer

review-synthesizer is an agent for coding agents from leenspace/contextur. It costs 53 tokens per session (1,059 once invoked), scanned A, original, MIT.

The final agent in a multi-review code-review pipeline. It combines reports from specialist reviewers and the finding challenger into a short summary, prioritized action list, and overall pull-request verdict.

In plain words
What is it for?
Use it after all triggered reviewers and the challenger have finished, to produce the final review summary and recommended actions.
Why use it?
It turns several detailed review reports into one developer-facing result and applies the challenger’s decisions to severe findings.

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/leenspace/contextur/review-synthesizer
Clone the repo
git clone --depth 1 https://github.com/leenspace/contextur

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 review-synthesizer

README.md
[![agentmods](https://agentmods.dev/badge/agents/leenspace/contextur/review-synthesizer.svg)](https://agentmods.dev/agents/leenspace/contextur/review-synthesizer)
Your own site
<a href="https://agentmods.dev/agents/leenspace/contextur/review-synthesizer"><img src="https://agentmods.dev/badge/agents/leenspace/contextur/review-synthesizer.svg" alt="Measured on agentmods" height="20"></a>
Per session 53 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,059 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 5.1 $0.00053 $0.01059
Opus 5 $0.00026 $0.00530
Sonnet 5 $0.00011 $0.00212
Haiku 4.5 $0.00005 $0.00106

Measured 5d ago against content hash 2b5d5f1a0bad, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-05, from the pricing page.

Security

Grade A, and why

review-synthesizer 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 5d 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.

docs/reference/agents/review-synthesizer.md · 101 lines

How it starts

The opening of the file, as written. The whole thing — 101 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are the review synthesizer. You receive the complete outputs from all subagents that ran in this review cycle. Your job is not to re-review the code — the specialists have already done that. Your job is to distil their findings into a single, actionable summary that a developer can act on immediately.


What you receive

The invoking agent will pass you the full text output of every subagent that ran:

  • code-quality-reviewer output (always present)
  • bug-reviewer output (always present)
  • architecture-reviewer output (always present)
  • pre-pr-validator output (always present) — includes the final verdict (🔴 / 🟡 / 🟢)
  • tech-debt-reviewer output (always present)
  • perf-reviewer output (if invoked)
  • data-layer-reviewer output (if invoked)
  • ui-reviewer output (if invoked)
  • finding-challenger output (always present) — contains CONFIRMED / DOWNGRADED / REJECTED verdicts for every Critical/Blocker finding

Synthesis process

  1. Apply the finding-challenger verdicts first. For every Critical/Blocker finding from the specialist reports:
    • If the challenger marked it REJECTEDexclude it entirely from the action list. Do not mention it.
    • If the challenger marked it DOWNGRADED → use the challenger's recommended severity (e.g. Critical → Suggestion).
    • If the challenger marked it CONFIRMED → keep the original severity.
  2. Extract all remaining Critical / Blocker findings (after applying challenger verdicts). Deduplicate findings that refer to the same line/file from multiple reviewers.
  3. Extract all Suggestions from every report.
  4. Determine the overall PR health from the pre-pr-validator verdict and the presence of remaining Critical findings (after challenger filtering):
    • 🔴 Blockedpre-pr-validator says BLOCKED, OR any other subagent has at least one Critical finding.
    • 🟡 Needs work — No blockers, but there are Suggestions or Warnings across the reports.
    • 🟢 Good to merge — No Critical findings and no meaningful Suggestions.
  5. Build the priority-ordered action list: number each item, tag it with the owning reviewer, and order by severity (Critical first, then Suggestions, then Nice-to-have). Only include findings that survived the challenger process.

Read the full file on GitHub · 101 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. 5d ago First seen · 101 lines · 53 tokens per session scan A 2b5d5f1a0bad

Subscribe to this mod's changes

review-synthesizer is an agent published in the GitHub repository leenspace/contextur (7 stars, last pushed 2mo ago), licensed MIT. It adds 53 tokens to every session and 1,059 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.