Agent-friendly documentation: a Markdown index for AI agents is at /llms.txt. Every page is also available as clean Markdown by appending .md to its URL (e.g. /pricing.md), or by requesting it with an Accept: text/markdown header.

LinkForgeAI
FIELD NOTES

What We Learn Running AI Agents Against the Open Web

Engineering notes from the team building LinkForge — scoring, automation, verification, and the parts of link building that resist automation.

AllAuthority scoringAutomationVerificationEngineering
Authority scoring· 9 min read

Why We Stopped Buying Domain Authority and Computed Our Own

A rented quality metric makes every qualification decision unexplainable and every renewal a dependency. Here is the signal set we replaced it with, how the weights were chosen, and what it costs to be wrong.

AP
Anurag Pattnaik
July 21, 2026
LAS BREAKDOWN · publisher #48211
Index status18
Domain age11
Visibility14
Tech health13
Spam signals6
Category fit12
74/ 100 · qualified

Recent Posts

Automation
CAPTCHA Is a Product Decision, Not a Bug

We detect challenges and hand them to a human. That choice caps autonomous throughput — and it is the right trade for anyone who plans to keep their domains.

July 14, 2026 · 6 min read
Automation
Teaching an Agent to Read a Form It Has Never Seen

Field mapping starts as a similarity search against forms we already solved, and only falls back to inference. The embedding step pays for itself by the second campaign.

July 7, 2026 · 8 min read
Verification
Submitted Is Not Live: Building an Evidence Trail Clients Can Audit

Anchor text, follow status, indexation, screenshot, HTML snapshot, timestamp. What we store per check, and why re-verification schedules matter more than the first result.

June 29, 2026 · 7 min read
Engineering
Eight Agents, One Graph: Keeping Failure Local

A slow submission stage should never stall discovery. How a state machine per campaign lets every stage retry and resume without corrupting the run.

June 20, 2026 · 10 min read
Authority scoring
The Middle Band: Designing a Review Queue Operators Actually Use

Scores between 40 and 59 are not a scoring failure, they are a routing decision. Treating manual review as a first-class state changed the qualification machine.

June 11, 2026 · 5 min read
Engineering
Unique at Scale: Semantic De-duplication of Generated Copy

Five hundred descriptions in one category will collapse toward each other unless you measure it. The similarity threshold we settled on, and how we enforce it pre-write.

June 3, 2026 · 7 min read
One engineering note a month. Nothing else.

What we shipped, what broke, and what the numbers looked like afterward.