From 9732b9078a460f7da151dc6f56e2543981a7b5a9 Mon Sep 17 00:00:00 2001 From: Vladimir Kalnitsky Date: Tue, 12 Nov 2024 20:57:35 +0400 Subject: [PATCH] Update the docs --- docs/index.md | 26 ++++++++++++++++++++++++-- docs/tutorial.md | 24 ------------------------ mkdocs.yml | 2 +- 3 files changed, 25 insertions(+), 27 deletions(-) diff --git a/docs/index.md b/docs/index.md index 3d43da7..e0770ac 100644 --- a/docs/index.md +++ b/docs/index.md @@ -1,2 +1,24 @@ -- [Module system](./modules.md) -- [Syntax](./syntax.md) +# Overview + +Metaprompt is a language for prompt automation, structuring and reuse. + +One one side, it is very similar to a template engine like Jinja or EJS. The twist is, metaprompt expansion depends on LLM outputs of LLM queries, that are formed in natural language. + +# Use cases + +## Templating + +The most basic use case of metaprompt is substituting parameter values instead of variable names embedded in a prompt. + +## Prompt organization + +A module system and a package system allow anyone to assign identities to promps and package them as callable functions. + +## Meta-prompting + +Meta-prompting is a technique of asking an LLM to create/modify an LLM prompt. + +- Dynamically crafting task-specific prompts based on a set of high level principles +- Modifying prompts to increase accuracy +- Securing inputs from prompt injection attacks +- Selecting the most suitable model based on prompt contents diff --git a/docs/tutorial.md b/docs/tutorial.md index cde6662..e69de29 100644 --- a/docs/tutorial.md +++ b/docs/tutorial.md @@ -1,24 +0,0 @@ -# Overview - -Metaprompt is a language for prompt automation, structuring and reuse. - -One one side, it is very similar to a template engine like Jinja or EJS. The twist is, metaprompt expansion depends on LLM outputs of LLM queries, that are formed in natural language. - -# Use cases - -## Prompt organization - -A module system and a package system allow anyone to assign identities to promps and package them as callable functions. - -## Meta-prompting - -Meta-prompting is a technique of asking an LLM to create an LLM prompt. - -- Using prompts to craft task-specific prompts based on a set of high level principles -- Modifying prompts to increase accuracy -- Securing inputs from prompt injection attacks -- Model selection based on prompt contents - -## Templating - -The most basic use case of metaprompt is substituting parameter values instead of variable names embedded in a prompt. diff --git a/mkdocs.yml b/mkdocs.yml index 1aed7a9..348fd6e 100644 --- a/mkdocs.yml +++ b/mkdocs.yml @@ -8,7 +8,7 @@ site_url: "https://docs.metaprompt-lang.org" # Theme theme: - name: mkdocs + name: readthedocs markdown_extensions: - toc: