Prompt

👋 Sign in for the ability to sort posts by relevant, latest, or top.

Cover image for Multi-Shot vs Zero-Shot: When Adding Examples Actually Hurts Accuracy

Gabriel Anhaia

May 24

Multi-Shot vs Zero-Shot: When Adding Examples Actually Hurts Accuracy

#ai #llm #prompt #benchmark

8 min read

Matthew Gladding

May 22

Addressing Hallucinations and Security in Open-Source LLM Agents

#security #agents #prompt #hallucinations

2 min read

Cover image for 3 lines I deleted from my AI cold email prompt — and reply rates jumped 8x

bluefin ships

May 19

3 lines I deleted from my AI cold email prompt — and reply rates jumped 8x

#ai #prompt #productivity #marketing

4 min read

Jake Counsell

May 7

I made a totally free tool where you can post / save / remix public and private prompts and have them instantly scored with insight on how to improve them.

#ai #promptengineering #prompt

1 min read

Evan Li

May 5

5 Prompt Recipes That Reliably Produce Great Images on Nano Banana Pro

#ai #prompt #tutorial #imagegeneration

3 min read

Cover image for Spud Was the Rumored GPT-6. It Shipped as GPT-5.5, Two Tiers Inside.

Gabriel Anhaia

Apr 27

Spud Was the Rumored GPT-6. It Shipped as GPT-5.5, Two Tiers Inside.

#ai #llm #architecture #prompt

7 min read

Cover image for The JSON-Mode Prompt Pattern That Survives Claude Version Bumps

Gabriel Anhaia

Apr 26

The JSON-Mode Prompt Pattern That Survives Claude Version Bumps

#ai #llm #prompt #tutorial

7 min read

Cover image for The 2-Line Defense That Stops 90% of Real-World Prompt Injection

Gabriel Anhaia

Apr 26

The 2-Line Defense That Stops 90% of Real-World Prompt Injection

#ai #llm #security #prompt

7 min read

Cover image for The Single Unit Test Every LLM Prompt Should Have

Gabriel Anhaia

Apr 26

The Single Unit Test Every LLM Prompt Should Have

#ai #llm #prompt #tutorial

7 min read

Cover image for Anthropic Skills Is Quietly Killing the Prompt Management Category

Gabriel Anhaia

Apr 26

Anthropic Skills Is Quietly Killing the Prompt Management Category

#ai #llm #prompt #promptengineering

7 min read

Cover image for The Most Expensive Hallucination of 2026: A Court Filing Goes Sideways

Gabriel Anhaia

Apr 26

The Most Expensive Hallucination of 2026: A Court Filing Goes Sideways

#ai #llm #prompt #observability

8 min read

Cover image for The "Let's Think Step by Step" Trick Is Now Hurting Your Prompts. Here's What Replaced It.

Gabriel Anhaia

Apr 26

The "Let's Think Step by Step" Trick Is Now Hurting Your Prompts. Here's What Replaced It.

#ai #llm #prompt #promptengineering

7 min read

Cover image for GLM-5.1 Beat GPT-5.4 at Coding — Under MIT License. Should You Actually Switch?

Gabriel Anhaia

Apr 26

GLM-5.1 Beat GPT-5.4 at Coding — Under MIT License. Should You Actually Switch?

#ai #llm #programming #prompt

8 min read

Cover image for Multi-Turn Jailbreaks Are the New Prompt Injection

Gabriel Anhaia

Apr 26

Multi-Turn Jailbreaks Are the New Prompt Injection

#ai #llm #security #prompt

8 min read

Cover image for JSON Mode Still Fails. Here's the Recovery Ladder

Gabriel Anhaia

Apr 27

JSON Mode Still Fails. Here's the Recovery Ladder

#ai #llm #python #prompt

6 min read

👋 Sign in for the ability to sort posts by relevant, latest, or top.

Forem

# prompt

Multi-Shot vs Zero-Shot: When Adding Examples Actually Hurts Accuracy

Addressing Hallucinations and Security in Open-Source LLM Agents

3 lines I deleted from my AI cold email prompt — and reply rates jumped 8x

I made a totally free tool where you can post / save / remix public and private prompts and have them instantly scored with insight on how to improve them.

5 Prompt Recipes That Reliably Produce Great Images on Nano Banana Pro

Spud Was the Rumored GPT-6. It Shipped as GPT-5.5, Two Tiers Inside.

The JSON-Mode Prompt Pattern That Survives Claude Version Bumps

The 2-Line Defense That Stops 90% of Real-World Prompt Injection

The Single Unit Test Every LLM Prompt Should Have

Anthropic Skills Is Quietly Killing the Prompt Management Category

The Most Expensive Hallucination of 2026: A Court Filing Goes Sideways

The "Let's Think Step by Step" Trick Is Now Hurting Your Prompts. Here's What Replaced It.

GLM-5.1 Beat GPT-5.4 at Coding — Under MIT License. Should You Actually Switch?

Multi-Turn Jailbreaks Are the New Prompt Injection

JSON Mode Still Fails. Here's the Recovery Ladder