Delv
General Assistantby ChatPRD★ 3.7

ChatPRD

AI product manager for PMs that drafts PRDs, provides expert coaching and integrates with Notion, Linear and Jira.

Try it
C
Safety & Trust

Delv Safety Grade: C

Score 58/100 · assessed 2026-04-19

Maintainer45
Permissions55
Supply chain40
Transparency35
Incidents100

ChatPRD is a web-based autonomous agent marketed as an AI product manager. It operates through a structured interview workflow to generate PRDs and integrates with Notion, Linear, and Jira. The maintainer is a single commercial entity with no visible open-source presence or repository, making verification of security practices impossible. The service requires OAuth access to third-party platforms (Notion, Linear, Jira) for write operations, creating a broad permissions surface. Supply chain is opaque as there's no package distribution or visible dependency management. Transparency is poor with no public code, security documentation, or incident response process. The freemium model with paid tiers suggests ongoing maintenance, but the lack of public accountability mechanisms and closed-source nature present meaningful trust concerns for teams handling sensitive product strategy.

Green flags

  • Freemium model suggests ongoing commercial maintenance and support
  • Structured workflow reduces risk of unintended autonomous actions
  • Integration with established platforms (Notion, Linear, Jira) via OAuth
  • No known security incidents or takedowns since launch

Red flags

  • No public repository or source code available for security review
  • Requires OAuth write access to Notion, Linear, and Jira workspaces
  • Opaque data handling practices for PRD content and user conversations
  • Single commercial entity with no visible security certifications
  • No published security policy or incident response process

Permissions requested

Outbound networkIdentity writeRepo writeExternal LLM callSend messages
Assessed by Delv Editorial using public metadata. Grades are advisory and update as the ecosystem changes. They do not replace your own review of permissions and code before granting an agent access to sensitive systems.

Pricing

FREEMIUM — Free tier, paid from $5/mo

Platforms

web

Review

ChatPRD positions itself as an AI product manager, and the autonomy claim is half-true. It won't write a complete PRD from a vague prompt, but it does guide you through a structured interview process that feels closer to working with a junior PM than wrestling with ChatGPT. You answer questions about user problems, success metrics, edge cases, and it assembles a coherent spec. The workflow is: describe your feature idea, answer follow-up questions (usually 5-8 rounds), review the draft, then push it to Notion, Linear, or Jira. The Notion integration is the smoothest, creating a page with proper headings and acceptance criteria already formatted. Where it shines: drafting specs for straightforward features when you already know what you want but hate the blank-page problem. I used it for a notification settings redesign and it caught three edge cases I'd forgotten (what happens when a user has no email on file, how admins override defaults, timezone handling). The coaching mode is less useful than advertised. It offers generic product advice, not context-specific critique, so you're better off using the PRD drafter and ignoring the chat. Failure modes: it struggles with complex, multi-stakeholder features. I tried speccing a billing migration and it kept oversimplifying the payment provider logic. It also has no memory between sessions, so iterating on a draft means re-explaining context. The free tier limits you to three PRDs per month, which sounds generous until you realise most teams draft variations and throwaway specs constantly. Nearest competitor is Delibr, which offers similar PRD templates but requires more manual input. ChatPRD's interview format is faster for solo PMs who want to externalise their thinking. Delibr wins if you need version control and stakeholder comments baked in. ChatPRD also competes with just using Claude or GPT-4 with a good prompt, and honestly, for experienced PMs, a custom prompt library is cheaper and more flexible. ChatPRD's value is the structure it enforces, which matters most if you're new to product work or drowning in feature requests.
Verdict

Worth the free tier if you write 2-3 specs a month and want faster first drafts. The paid plan makes sense for junior PMs or founders doing product work part-time. Experienced PMs with established workflows will find it limiting.

Good at

  • Interview format surfaces edge cases you'd otherwise miss
  • Notion integration creates properly formatted specs in seconds
  • Faster than blank-page drafting for straightforward features
  • Acceptance criteria are specific and testable, not vague

Watch out

  • No session memory, so iterating on drafts means re-explaining context
  • Struggles with complex, multi-system features
  • Coaching mode offers generic advice, not tailored critique
  • Free tier's three PRDs per month runs out quickly
  • Experienced PMs can replicate this with a good prompt library

Use cases

  • PRD drafting
  • product coaching
  • spec writing