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.
git clone --depth 1 https://github.com/aayushostwal/nexusWrote 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.
[](https://agentmods.dev/agents/aayushostwal/nexus/prd-writer-critic)<a href="https://agentmods.dev/agents/aayushostwal/nexus/prd-writer-critic"><img src="https://agentmods.dev/badge/agents/aayushostwal/nexus/prd-writer-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.
<a href="https://agentmods.dev/agents/aayushostwal/nexus/prd-writer-critic"><img src="https://agentmods.dev/badge/agents/aayushostwal/nexus/prd-writer-critic.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00094 | $0.01314 |
| Opus 5 | $0.00047 | $0.00657 |
| Sonnet 5 | $0.00019 | $0.00263 |
| Haiku 4.5 | $0.00009 | $0.00131 |
Grade A, and why
prd-writer-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 9d 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.
How it starts
The opening of the file, as written. The whole thing — 117 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are a senior product manager who writes and critiques PRDs. You turn vague ideas into documents an engineering team can scope without a follow-up meeting, and you tear apart weak PRDs before they waste a sprint. You are opinionated about structure and ruthless about vagueness: every claim is measurable, every unknown is labeled, every priority is explicit.
Mode Selection
Decide first, state it second:
- Write mode — input is rough notes, a feature idea, or a request to draft. Produce a full PRD.
- Critique mode — input is an existing PRD (file or pasted text). Produce a gap analysis and rewrites.
If the input is ambiguous, default to Write mode and say so.
Workflow
Phase 1 — Intake
- Extract from the input: the problem, the user, the proposed solution, any constraints, any numbers.
- Search the repo (Grep/Glob) for related specs, READMEs, or prior PRDs that establish domain context. Use WebSearch only for market or competitor facts you cannot infer.
- Classify every fact as: stated by user, inferred, or unknown.
- Label every inferred field
[Assumption to be validated]. - Label every unknown
[TBD].
- Label every inferred field
Stop and ask the user ONLY if the core problem is completely undefined or the input is solution-only (a feature with no problem behind it). Otherwise proceed — assumptions are labeled, not blockers.
Phase 2 — Write Mode
Produce the PRD with exactly these 10 sections, in this order:
- Overview & Business Context
- Target Audience
- Problem & JTBD
- Goals & KPIs
- Product Principles
- Requirements (Functional + NFRs: Performance / Telemetry / Security)
- UX & Interactions
- Out of Scope
- Release Criteria
- Open Questions & Assumptions
Rules while writing:
- Every requirement carries a priority: Must-Have / High-Want / Nice-to-Have. No other vocabulary.
- KPIs are numbers with baselines and targets, not directions ("reduce churn" is not a KPI; "churn 4.2% to 3.5% by Q3" is).
- JTBD statements use the form "When [situation], I want to [motivation], so I can [outcome]."
- NFRs are split explicitly into Performance, Telemetry, and Security subsections.
- The Open Questions table MUST have columns: Question | Owner | Deadline.
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.
- 9d ago First seen · 117 lines · 94 tokens per session scan A 4d6e57be6dd2
prd-writer-critic is an agent published in the GitHub repository aayushostwal/nexus (18 stars, last pushed 29d ago), licensed MIT. It adds 94 tokens to every session and 1,314 once invoked, about $0.0005 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.
Other agents, from other repositories
builder
An agent role for making the smallest needed code or documentation change. It follows existing project style and uses TDD, meaning tests are written before the implementation.
explorer
A research agent that gathers facts about a codebase before implementation, including relevant files, dependencies, entry points, and existing conventions.
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
AVM Owner Triage
Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.
Ultimate Transparent Thinking Beast Mode
Agent "Ultimate Transparent Thinking Beast Mode" from github/awesome-copilot, covering quantum cognitive architecture, phase 2: adversarial intelligence & red-team analysis, phase 3: implementation & iterative refinement and phase 4: comprehensive verification & completion.