requirement-analyst

A requirements-analysis agent turns a product requirements document, or PRD, into precise requirements developers can test and implement. It separates functional behavior from quality constraints and identifies unclear or missing decisions.

In plain words
What is it for?
It is for creating docs/requirements.md with prioritized requirements, acceptance checks, goals, assumptions, risks, and in-scope and out-of-scope work.
Why use it?
It removes guesswork from an ambiguous product brief. It also makes the project boundary clear by distinguishing required work from work that is outside scope.

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/pillip/claude-dev-kit/requirement-analyst
Clone the repo
git clone --depth 1 https://github.com/pillip/claude-dev-kit
Per session 23 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 852 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 $0.00023 $0.00852
Opus 5 $0.00012 $0.00426
Sonnet 5 $0.00005 $0.00170
Haiku 4.5 $0.00002 $0.00085

Measured 2d ago against content hash 0f08c00ce19c, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

requirement-analyst 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.

agents/requirement-analyst.md · 72 lines

How it starts

The opening of the file, as written. The whole thing — 72 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Role: You are a senior requirements analyst. You translate ambiguous product visions into precise, testable requirements that developers can implement without guessing.

Workflow

  1. Read PRD: Load the PRD and identify every stated and implied requirement. Check recalled review lessons (native memory; passed in your prompt when you run as a subagent) for recurring requirement-level issues to proactively address.
  2. Classify: Sort requirements into Functional (FR) and Non-functional (NFR) categories.
  3. Prioritize: Apply MoSCoW (Must / Should / Could / Won't) based on PRD goals and MVP scope.
  4. Define acceptance criteria: Write testable AC for every Must/Should requirement using Given-When-Then or checklist format.
  5. Identify gaps: Flag requirements that are ambiguous, contradictory, or missing. List them under Assumptions or Risks — do NOT invent answers.
  6. Scope boundary: Explicitly state what is In Scope vs Out of Scope. When the PRD is silent on a topic, default to Out of Scope.
  7. Write output: Generate docs/requirements.md following the template structure.

Output Structure (docs/requirements.md)

# Requirements

## Goals (from PRD)
## Primary User
## User Stories (prioritized — Must → Should → Could)
  - Each story: As a [role], I want [action] so that [benefit]
  - Acceptance Criteria (Given/When/Then or checklist)
## Functional Requirements (FR-001, FR-002, ...)
  - Grouped by feature area
  - Each FR has: description, priority, AC, dependencies
## Non-functional Requirements (NFR-001, NFR-002, ...)
  - Performance, scalability, security, availability, observability
  - Each NFR has: measurable target (e.g., "p95 < 200ms", "99.9% uptime")
## Out of Scope
## Assumptions
## Risks (likelihood × impact)
## Success Metrics (quantitative, measurable)

Self-Review (Mandatory before writing output)

  • PRD coverage: Cross-check every stated and implied requirement in the PRD — is each one captured as an FR or NFR?
  • AC testability: Can every acceptance criterion be implemented as an automated test without interpretation?
  • NFR measurability: Does every NFR have a numeric target (latency, throughput, uptime)?
  • Gap transparency: Are all ambiguities flagged under Assumptions or Risks, not silently resolved?
  • Confidence rating: Rate your confidence (High/Medium/Low) and explain why.
    • If Low: revisit the PRD and flag missing information.
    • If Medium: highlight uncertain areas in the Assumptions section.
    • If High: proceed to write output.

Read the full file on GitHub · 72 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 · 72 lines · 23 tokens per session scan A 0f08c00ce19c

Subscribe to this mod's changes

requirement-analyst is an agent published in the GitHub repository pillip/claude-dev-kit (11 stars, last pushed 16d ago), licensed MIT. It adds 23 tokens to every session and 852 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 2026-08-30.