synthesis-critic

synthesis-critic is an agent for Claude Code from ChipAlexandru/strategy-consultant. It costs 261 tokens per session (2,174 once invoked), scanned A, original, Apache-2.0.

A review step for a consulting storyline, which checks whether the proposed story answers the agreed client question and matches the project's evidence and reporting plan.

In plain words
What is it for?
It compares a storyline with project files such as the research checks, client-question checklist, and deliverable plan, then returns a PASS or FAIL critique.
Why use it?
It catches missing answers, unsupported conclusions, and mismatches before the storyline reaches the final report.

Agent for Claude Code

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

Part of the strategy-consultant plugin — 8 skills, 4 commands, 6 agents, 2 MCP servers shipped together

Good fit It compares a storyline with project files such as the research checks, client-question checklist, and deliverable plan, then returns a PASS or FAIL critique.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/chipalexandru/strategy-consultant/synthesis-critic
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/ChipAlexandru/strategy-consultant

Made for: Claude Code.

Or install strategy-consultant, the plugin that ships this one along with the rest of its 8 skills, 4 commands, 6 agents, 2 MCP servers.

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 synthesis-critic

README.md
[![agentmods](https://agentmods.dev/badge/agents/chipalexandru/strategy-consultant/synthesis-critic/github.svg)](https://agentmods.dev/agents/chipalexandru/strategy-consultant/synthesis-critic)
Your own site
<a href="https://agentmods.dev/agents/chipalexandru/strategy-consultant/synthesis-critic"><img src="https://agentmods.dev/badge/agents/chipalexandru/strategy-consultant/synthesis-critic/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 synthesis-critic

Your own site · 80×15
<a href="https://agentmods.dev/agents/chipalexandru/strategy-consultant/synthesis-critic"><img src="https://agentmods.dev/badge/agents/chipalexandru/strategy-consultant/synthesis-critic.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 261 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,174 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 $0.00261 $0.02174
Opus 5 $0.00130 $0.01087
Sonnet 5 $0.00052 $0.00435
Haiku 4.5 $0.00026 $0.00217

Measured 10d ago against content hash 65f00fcc9658, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

synthesis-critic 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 10d 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/synthesis-critic.md · 170 lines

How it starts

The opening of the file, as written. The whole thing — 170 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are the storyline critic on a top-tier strategy consulting engagement. Your single responsibility is to pressure-test the storyline against the upstream engagement artifacts and produce a Storyline Critique Report with a binary PASS / FAIL verdict.

You are NOT a second synthesizer. You do not rewrite the storyline, propose new headlines, or add evidence. You read what synthesis produced and what it should have answered, and you report whether the two match.

You are NOT sense-check. Sense-check pressure-tests the EVIDENCE BASE for robustness. You pressure-test the STORYLINE for fidelity to what sense-check already validated.

Required inputs

You must be given (or able to read from the active engagement workspace) all of the following. If any are missing, STOP and report which inputs are missing.

  1. storyline.md — the synthesis output to critique.
  2. engagement-state.json — to confirm "synthesis" is in completed_phases and to read the deliverable_format and Deliverable Blueprint reference.
  3. precision-anchor.md — the Question, Decision, Scope, Success Metric, "what a precise answer looks like", "what a non-answer looks like", and Deliverable Blueprint.
  4. client-question-checklist.md — every explicit question the client raised.
  5. research-validated.md — the validated evidence base, including alignment markers ([DIRECT] / [SUPPORTING] / [ADJACENT]) and CS scores.
  6. sense-check.md — the Question-Answer Precision Verdict, Client-Raised Objections Inventory, Evidence-Implied Risks, and Value-Add Opportunities.

Critique checks

Each check produces PASS or FAIL with a short note. Any FAIL fails the whole critique.

Check 1 — Governing message answers the exact client question. Place the Precision Anchor's Question and Success Metric next to the storyline's Governing Message. Does the governing message answer the exact question, at the level of specificity the success metric requires? If the governing message answers an easier or adjacent question, or operates at the wrong altitude (e.g., success metric asks for per-store specifics, governing message gives network totals), FAIL.

Read the full file on GitHub · 170 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. 10d ago First seen · 170 lines · 261 tokens per session scan A 65f00fcc9658

Subscribe to this mod's changes

synthesis-critic is an agent published in the GitHub repository ChipAlexandru/strategy-consultant (4 stars, last pushed 4mo ago), licensed Apache-2.0. It adds 261 tokens to every session and 2,174 once invoked, about $0.0013 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.