AI system · 2026
Content research agent
Turns an idea into a researched article and a post for each channel. Every fact traces to a page it read, and nothing goes out without approval.
- Role
- Design, build, deploy
- Stack
- Next.js, Supabase, pgvector, OpenAI embeddings, Firecrawl, Resend
- Status
- Archived
Overview
A content manager needs a researched article and a post for every channel, with every fact traceable to a page the system actually read.
Decisions
A person picks the sources and the angle.
Before anything is written. Only the chosen angle gets written, so the expensive step runs once.
Facts are checked in code.
Sources, facts and format are checked before the grader sees the draft. The model only judges tone and clarity.
Links come from the sources.
The model never writes a link, and nothing publishes before a person approves it. The server enforces that.
Architecture
- Person
- Tool
- Model
- Check
- Result
Step 1, Human: An idea or a link
Content manager
Step 2, Step: Read the sources
Search, Firecrawl
Step 3, AI: Three angles
Haiku
Step 4, Human: Pick the sources and angle
Content manager
Step 5, AI: Write the article
Sonnet 5
Step 6, Logic: Passes the grade?
Code, Sonnet 5
Then: step 5 (no: revise); step 7 (yes).
Step 7, AI: A version per channel
Haiku
Step 8, Human: Approve each channel
Content manager
Step 9, Output: Released on schedule
Resend
Results
41 / 44
claims traced to a source
4:54
minutes from idea to graded article
$0.24
for that full run
From the test report.
Known limits
The grader isn't benchmarked.
I moved it to a cheaper model without testing the two side by side.
Revisions don't always land.
Two rounds didn't always fix the writing, and one test article still needed a person.