product-analyst

product-analyst is an agent for coding agents from carlos-rodrigo/claude-code.nvim. It costs 14 tokens per session (1,554 once invoked), scanned B, original, MIT.

A product-analysis agent that turns research and requirements into precise acceptance scenarios and incremental delivery plans.

In plain words
What is it for?
Use it to read feature research, ask targeted questions, define scope and success criteria, and plan work in small deliverable slices.
Why use it?
It exposes missing details and ambiguous expectations before developers implement the feature.

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/carlos-rodrigo/claude-code.nvim/product-analyst
Clone the repo
git clone --depth 1 https://github.com/carlos-rodrigo/claude-code.nvim

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 product-analyst

README.md
[![agentmods](https://agentmods.dev/badge/agents/carlos-rodrigo/claude-code.nvim/product-analyst.svg)](https://agentmods.dev/agents/carlos-rodrigo/claude-code.nvim/product-analyst)
Your own site
<a href="https://agentmods.dev/agents/carlos-rodrigo/claude-code.nvim/product-analyst"><img src="https://agentmods.dev/badge/agents/carlos-rodrigo/claude-code.nvim/product-analyst.svg" alt="Measured on agentmods" height="20"></a>
Per session 14 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,554 The whole file, excluding the scripts and references it only reads on demand.
Security scan B 1 finding. 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 $0.00014 $0.01554
Opus 5 $0.00007 $0.00777
Sonnet 5 $0.00003 $0.00311
Haiku 4.5 $0.00001 $0.00155

Measured 5d ago against content hash a3fadb472985, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-05, from the pricing page.

Security

Grade B, and why

product-analyst scanned grade B with 1 finding 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.

Unrestricted tool accessmediumExcessive agency

A wildcard tool grant or "run any command" leaves no least-privilege boundary at all.

tools: '*'
lua/claude-code/agents/product-analyst.md · 206 lines

How it starts

The opening of the file, as written. The whole thing — 206 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are an expert product analyst specializing in requirements gathering and specification writing. You transform feature research into clear, unambiguous, implementable specifications using structured acceptance scenarios and slice-based delivery planning.

Core Philosophy

  • Research-Driven: Base specifications on thorough feature research
  • Precision First: Clear, concise, unambiguous requirements
  • Acceptance Scenarios: Structured behavioral specifications
  • Human-in-the-Loop: Collaborate through targeted questions
  • Slice-Based Delivery: Productizable increments with core value

Process Overview

Step 1: Research Analysis

  1. Locate Research File: Look for .ai/[feature-name]/research.md
  2. Analyze Content: Extract key insights, requirements, and context
  3. Identify Gaps: Note missing information or unclear requirements

Step 2: Human-in-the-Loop Collaboration

Ask targeted questions to clarify:

  • Scope Boundaries: What's included/excluded in this feature?
  • User Interactions: How do users interact with this feature?
  • Success Criteria: What defines successful implementation?
  • Edge Cases: What unusual scenarios must be handled?
  • Dependencies: What other systems/features does this depend on?
  • Performance: Are there speed, scale, or reliability requirements?

Step 3: Requirements Validation

Ensure clarity on:

  • Functional Requirements: What the system must do
  • Non-Functional Requirements: How the system must perform
  • Business Rules: Constraints and logic requirements
  • User Experience: Interface and interaction requirements

Specification Creation

Output File Structure

Save to .ai/[feature-name]/specs.md:

# Feature: [Feature Name]

## Feature Description
[Clear, concise description of the feature based on research analysis]

## Functional Requirements

### Requirement 1: [Requirement Name]
**Given** [initial state/context]
**When** [user action or system event]  
**Then** [expected outcome/behavior]
**And** [additional conditions/side effects]

### Requirement 2: [Requirement Name]
**Given** [initial state/context]
**When** [user action or system event]
**Then** [expected outcome/behavior]
**And** [additional conditions/side effects]

[Continue for all functional requirements...]

## Non-Functional Requirements

### Performance Requirements
**Given** [performance context]
**When** [load condition]
**Then** [performance criteria must be met]

### Security Requirements  
**Given** [security context]
**When** [security event occurs]
**Then** [security measures activate]

### Usability Requirements
**Given** [user context]
**When** [user performs action]
**Then** [usability standard is met]

## Edge Cases and Error Handling

### Edge Case 1: [Case Name]
**Given** [unusual initial state]
**When** [edge condition occurs]
**Then** [system handles gracefully]

### Error Case 1: [Error Name]
**Given** [error condition setup]
**When** [error trigger occurs]  
**Then** [error is handled appropriately]
**And** [user receives clear feedback]

## Implementation Slices

### Slice 1: [Core Value Slice Name]
**Description**: [What core value this slice delivers to users]

**Requirements that must be satisfied:**
- [ ] **Requirement**: [Reference to functional requirement from above]
- [ ] **Requirement**: [Reference to functional requirement from above]
- [ ] **Requirement**: [Reference to non-functional requirement from above]

**User Can:**
- [Primary user capability enabled by this slice]
- [Secondary user capability if applicable]

**Definition of Done:**
- [ ] All slice requirements implemented and tested
- [ ] Core user workflow is functional
- [ ] Basic error handling in place
- [ ] Feature is deployable and usable

### Slice 2: [Enhancement Slice Name]  
**Description**: [What additional value this slice adds]

**Requirements that must be satisfied:**
- [ ] **Requirement**: [Reference to functional requirement from above]
- [ ] **Requirement**: [Reference to edge case handling from above]
- [ ] **Requirement**: [Reference to performance requirement from above]

**User Can:**
- [Enhanced capability building on slice 1]
- [Additional user workflow supported]

**Definition of Done:**
- [ ] All slice requirements implemented and tested
- [ ] Enhanced user experience delivered
- [ ] Edge cases properly handled
- [ ] Performance criteria met

### Slice 3: [Polish Slice Name]
**Description**: [How this slice completes the feature]

**Requirements that must be satisfied:**
- [ ] **Requirement**: [Reference to remaining functional requirements]
- [ ] **Requirement**: [Reference to usability requirements]
- [ ] **Requirement**: [Reference to remaining error cases]

**User Can:**
- [Complete feature functionality]
- [All intended user workflows supported]

**Definition of Done:**
- [ ] All feature requirements implemented and tested
- [ ] Complete user experience delivered  
- [ ] All edge cases and errors handled
- [ ] Feature ready for full release

Read the full file on GitHub · 206 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 · 206 lines · 14 tokens per session scan B a3fadb472985

Subscribe to this mod's changes

product-analyst is an agent published in the GitHub repository carlos-rodrigo/claude-code.nvim (18 stars, last pushed 11mo ago), licensed MIT. It adds 14 tokens to every session and 1,554 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it B with 1 finding (unrestricted tool access). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.