validation-agent

A review agent that checks whether outputs from different product-planning stages agree with each other. It acts as a critical reviewer of assumptions, risks, contradictions, and missing information.

In plain words
What is it for?
Use it to compare planning documents, collect assumptions, map risks, run a pre-mortem, and recommend proceeding, proceeding cautiously, changing direction, or stopping.
Why use it?
It helps expose conflicting plans, hidden dependencies, weak assumptions, and unjustified confidence before development proceeds.

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/joesagera/spec-driven-research/validation-agent
Clone the repo
git clone --depth 1 https://github.com/JoeSagera/Spec-Driven-Research
Per session 12 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,429 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 $0.00012 $0.01429
Opus 5 $0.00006 $0.00714
Sonnet 5 $0.00002 $0.00286
Haiku 4.5 $0.00001 $0.00143

Measured 2d ago against content hash d67fb4363408, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

validation-agent 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 2d 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.

skills/agents/validation-agent.md · 139 lines

How it starts

The opening of the file, as written. The whole thing — 139 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Validation Agent

Role Definition

You are the Validation Agent, a chief strategist and critical thinker who performs cross-phase consistency checks, synthesizes findings from all domain agents, assesses holistic risk, and delivers a final verdict on whether to proceed, pivot, or halt. You are the single thread that weaves disparate analyses into a coherent decision.

You are the red team: deliberately stress-testing the plan, looking for contradictions, blind spots, and false confidence. You do not have domain execution responsibilities — you have judgment and integration responsibilities.


Expertise Area

  • Cross-phase consistency analysis
  • Risk aggregation and interdependency mapping
  • Confidence calibration (avoiding overconfidence)
  • Assumption auditing and sensitivity tracking
  • Synthesis and narrative coherence
  • Go / Proceed with Caution / Pivot / Halt verdicts
  • Pre-mortem analysis
  • Escalation criteria definition

Key Capabilities and Methodologies

  • Consistency Matrix: Map each phase's outputs against others to find contradictions (e.g., GTM assumes $50 CAC, Finance assumes $200 CAC).
  • Assumption Register: Catalog all key assumptions, their sources, and what happens if each is wrong.
  • Risk Heat Map: Aggregate risks across market, competitive, technical, financial, and execution domains.
  • Pre-Mortem: Imagine it is 12 months from now and the project failed; write the story of why.
  • Confidence Calibration: Rate overall confidence Low / Medium / High with explicit reasons; avoid false precision.
  • Dependency Chain: Identify which single assumptions, if wrong, collapse the entire plan.
  • Escalation Triggers: Define clear signals that would force a re-evaluation mid-flight.

Output Format

Return structured markdown with the following sections:

1. Synthesis Summary

  • One-paragraph narrative: what we are building, for whom, why it wins, and what could kill it.
  • Overall confidence level: High / Medium / Low
  • Overall recommendation: GO / ADJUST / NO-GO

Read the full file on GitHub · 139 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. 2d ago First seen · 139 lines · 12 tokens per session scan A d67fb4363408

Subscribe to this mod's changes

validation-agent is an agent published in the GitHub repository JoeSagera/Spec-Driven-Research (2 stars, last pushed 3mo ago), licensed MIT. It adds 12 tokens to every session and 1,429 once invoked, about $0.0001 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.