Pipeline
End-to-End Agent Pipeline
Researcher, Writer, Fact-Checker, Style Polisher, and Rubric Grader run as a coordinated workflow instead of a single prompt.
Core Capabilities
Built for Traceable Content Ops
Clear ownership, measurable quality, and auditable execution from prompt input to final output.
Security
Auth + Ownership
Supabase Auth secures access. Runs are tied to a user and protected by ownership checks and RLS policies.
Quality
Quality Gate + Rollback
Rubric scoring enforces clarity, correctness, and completeness. Failed runs can automatically retry from earlier stages.
Observability
Operational Visibility
Track status, logs, token usage, latency, durations, and rubric outcomes across the runs and dashboard surfaces.
Storage
Dual Persistence
Local JSON run state keeps iteration simple while Supabase Postgres stores analytics-grade run, step, and LLM call records.
Distribution
Publishing Extensions
After the core blog pipeline, generate LinkedIn pack artifacts and optional images from the same run context.
Workflow
How It Works
A structured, inspectable pipeline from authenticated input to publish-ready output.
01
Access
Authenticate
Sign in with Supabase Auth. Every run is tied to your user and protected by ownership checks.
02
Input
Create a Run
In Studio, submit PRD, tone, audience, model route, and optional web-search settings.
03
Execute
Pipeline A Executes
Research -> Writer -> Fact-Checker loop -> Style Polisher execute with per-step timeline logs.
04
Gate
Quality Gate Applies
Rubric scoring checks clarity, correctness, and completeness. Failed outputs can rollback and retry.
05
Observe
Monitor Reliability
Use /runs for per-run detail and /dashboard for aggregate token, latency, and quality metrics.
06
Deliver
Ship and Extend
Export the final blog, then optionally run LinkedIn pack and image generation downstream.
Pricing
Plans Built Around Usage Caps
Main pricing is subscription-based with run and token caps. This keeps usage predictable while aligning with token-based LLM costs.
Free (Trial)
Trial$0
/month · BDT 0
- 3 runs per month
- No web research (or very limited)
- Basic blog output only
Starter (Solo)
Popular$9
/month · BDT 999
- 30 runs per month
- Token budget cap: 300k tokens/month
- Basic logs and export
Pro (Creator)
Best Value$19
/month · BDT 1,999
- 120 runs per month
- Token budget cap: 1M tokens/month
- Web research enabled
- Rubric quality gate + retry/rollback
- Full audit logs + dashboard
Credit Pack Add-ons
For users who do not want subscriptions, offer one-time run top-ups.
One-time purchase
Run Pack 50
$10
BDT 1,000
One-time top-up for extra usage.
Run Pack 200
$30
BDT 3,000
One-time top-up for power users.
Ready to Start
Interested in shipping your next editorial workflow faster?
Launch a complete PRD-to-blog run with citations, quality gates, and metrics in one guided flow.