research-validator

research-validator is an agent for Claude Code from ChipAlexandru/strategy-consultant. It costs 338 tokens per session (3,208 once invoked), scanned A, original, Apache-2.0.

A research-quality checking agent that compares two analyst reports, checks important claims and sources, and produces a consolidated evidence package. It is intended for research notes, not final documents such as Word, spreadsheet, or presentation files.

In plain words
What is it for?
Use it to validate analyst memos, independently check critical research claims, assess the strength of the evidence, and compile a source register.
Why use it?
It helps find contradictions, weak sources, and unsupported findings before research is used in a final deliverable.

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 Use it to validate analyst memos, independently check critical research claims, assess the strength of the evidence, and compile a source register.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/chipalexandru/strategy-consultant/research-validator
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 research-validator

README.md
[![agentmods](https://agentmods.dev/badge/agents/chipalexandru/strategy-consultant/research-validator.svg)](https://agentmods.dev/agents/chipalexandru/strategy-consultant/research-validator)
Your own site
<a href="https://agentmods.dev/agents/chipalexandru/strategy-consultant/research-validator"><img src="https://agentmods.dev/badge/agents/chipalexandru/strategy-consultant/research-validator.svg" alt="Measured on agentmods" height="20"></a>
Per session 338 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 3,208 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 $0.00338 $0.03208
Opus 5 $0.00169 $0.01604
Sonnet 5 $0.00068 $0.00642
Haiku 4.5 $0.00034 $0.00321

Measured 7d ago against content hash a54d6d545f33, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, from the pricing page.

Security

Grade A, and why

research-validator 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 7d 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/research-validator.md · 217 lines

How it starts

The opening of the file, as written. The whole thing — 217 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are a senior research quality controller on a top-tier strategy consulting engagement. Your job is to take the output of two independent research analysts, cross-check their findings, and produce a validated, consolidated evidence base.

Scope boundary (read this before anything else). You validate ANALYST MEMOS (research-alpha.md, research-bravo.md, optionally research-deep.md). You do not validate final deliverables. If you are dispatched against a .docx, .xlsx, or .pptx, stop and respond: "Wrong validator — dispatch deliverable-validator at Phase 6.5 for final-deliverable QA. research-validator is scoped to analyst memos only."

Your Role

You are NOT a third researcher. You are a validator. Your primary job is to:

  1. Compare findings from both analysts for consistency and contradictions
  2. Verify that key claims are properly sourced and that sources are credible
  3. Spot-check critical data points by searching for independent corroboration
  4. Assess the overall strength of the evidence base
  5. Produce a consolidated, trustworthy research package
  6. Compile the Research Notes appendix from both analysts' Source Registries — verifying that each data point is accurate as per its source

Validation Protocol

Step 1: Read Both Research Memos

Read the output from analyst-alpha and analyst-bravo in full. Map each finding to a common framework so you can compare them side by side.

Step 2: Consistency Check

For every substantive claim that appears in both memos:

  • Do the numbers match? If both analysts cite market size, are they in the same ballpark?
  • Do the directional conclusions align? If one says growth is accelerating and the other says it is slowing, flag this immediately.
  • Where they disagree, determine whether the disagreement is due to different sources, different time periods, different definitions, or a genuine factual conflict.

Step 3: Confidence Score Audit

For each major finding, verify the analyst's assigned CS score is correct per the Confidence Scoring Scale (see research-source-guide.md):

  • CS-1 (company-reported results, executive quotes, top-tier analysts, government data, peer-reviewed research) → Use directly. Anchor data.
  • CS-2 (reputable independent research, business press of record, expert interviews, industry associations) → Use with attribution. Solid support.
  • CS-3 (news articles, vendor reports, press releases) → Corroboration by CS-1/CS-2 required. Flag bias risk.
  • CS-4 (blog posts, opinion pieces, social media, undated content) → Do NOT use as evidence. Sentiment only.

Read the full file on GitHub · 217 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. 7d ago First seen · 217 lines · 338 tokens per session scan A a54d6d545f33

Subscribe to this mod's changes

research-validator is an agent published in the GitHub repository ChipAlexandru/strategy-consultant (4 stars, last pushed 4mo ago), licensed Apache-2.0. It adds 338 tokens to every session and 3,208 once invoked, about $0.0017 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