Test prompts with ollama models

Author: slimslenderslacks

prompts/dockerfiles_llama3.1/010_system_prompt.md

@@ -0,0 +1,2 @@
+You are an assistant who specializes in authoring Dockerfiles for projects.
+Since you are an expert and know about their project, be definitive about recommendations.

prompts/dockerfiles_llama3.1/100_user_prompt.md

@@ -0,0 +1,21 @@
+Write Dockerfiles using three stages.  Here are the three stages to use:
+1. the first stage should be called "deps" 
+   and it should fetch the runtime dependencies using npm ci
+   `with the --omit=dev` flag.
+2. The second stage should be called build
+   and it should be based on the deps stage. 
+   It should run npm ci first and then it should run npm build
+3. The third stage should start with a FROM line that uses the `node` base image. 
+   This stage should do then do three things.
+     1. it copies the node_modules directory from the deps stage.
+     2. it copies the dist directory from the build stage.
+     3. it should then create an entrypoint that runs npm start
+
+If you need to use a RUN statement containing `npm ci` always 
+add the argument `--mount=type=cache,target=/root/.npm` to the RUN instruction.  
+The `--mount` argument should be placed between the word RUN and the npm command.
+This will cache the npm packages in the docker build cache and speed up the build process.
+
+* Generate a Dockerfile for an NPM project.
+* Call a function to get the recommended tag for repository `node`
+* In each FROM line in the Dockerfile, use the recommended tag.

prompts/dockerfiles_llama3.1/README.md

@@ -0,0 +1,24 @@
+---
+extractors:
+  - name: go-linguist
+tool_choice: auto
+model: gpt-4
+stream: true
+functions:
+  - name: docker_scout_tag_recommendation
+---
+
+# Background
+
+These prompts add Dockerfile authoring skills to the assistant.
+
+* the [user prompts](100_user_prompt.md) direct the assist to extract details about the project, and then write a Dockerfile to the root of the project.
+* the [npm best practices](npm-best-practices.md) is are added to the system prompts whenever the assist detects that this is an NPM project.
+
+## functions
+
+These instructions in these prompts rely on 3 functions
+
+1. `analyze_project` - requires read-only access to the project and extracts details from the project to give the assistant context for authoring the Dockerfile
+2. `write_files` - has read-write access to the project and will write Dockerfiles so that developers don't have to copy and paste data out of a chat.
+3. `docker_scout_tag_recommendation` - uses Docker Scout to improve the choice tag selection during code generation.

prompts/dockerfiles_mistral-nemo/010_system_prompt.md

@@ -0,0 +1,2 @@
+You are an assistant who specializes in authoring Dockerfiles for projects.
+Since you are an expert and know about their project, be definitive about recommendations.

prompts/dockerfiles_mistral-nemo/100_user_prompt.md

@@ -0,0 +1,19 @@
+Write Dockerfiles for NPM projects using three stages.  Do these three steps sequentially.
+* the first node depemdencies stage should be called "deps" 
+  and it should fetch the runtime dependencies using npm ci
+  `with the --omit=dev` flag.
+* The second Dockerfile stage should be called build
+  and it should be based on the deps stage. 
+  It should run npm ci and then npm build
+* The third Dockerfile stage should select a base image 
+  that is recommended by Scout and use that in the FROM line.
+  This stage should do then do three things.
+    1. it copies the node_modules directory from the deps stage.
+    2. it copies the dist directory from the build stage.
+    3. it then runs npm start
+
+If you need to use a RUN statement containing `npm ci` always 
+add the argument `--mount=type=cache,target=/root/.npm` to the RUN instruction.  
+The `--mount` argument should be placed between the word RUN and the npm command.
+This will cache the npm packages in the docker build cache and speed up the build process.
+

prompts/dockerfiles_mistral-nemo/README.md

@@ -0,0 +1,24 @@
+---
+extractors:
+  - name: go-linguist
+tool_choice: auto
+model: gpt-4
+stream: true
+functions:
+  - name: docker_scout_tag_recommendation
+---
+
+# Background
+
+These prompts add Dockerfile authoring skills to the assistant.
+
+* the [user prompts](100_user_prompt.md) direct the assist to extract details about the project, and then write a Dockerfile to the root of the project.
+* the [npm best practices](npm-best-practices.md) is are added to the system prompts whenever the assist detects that this is an NPM project.
+
+## functions
+
+These instructions in these prompts rely on 3 functions
+
+1. `analyze_project` - requires read-only access to the project and extracts details from the project to give the assistant context for authoring the Dockerfile
+2. `write_files` - has read-write access to the project and will write Dockerfiles so that developers don't have to copy and paste data out of a chat.
+3. `docker_scout_tag_recommendation` - uses Docker Scout to improve the choice tag selection during code generation.

prompts/recommended_tags/010_system_prompt.md

@@ -0,0 +1 @@
+You are a Docker expert.

prompts/recommended_tags/100_user_prompt.md

@@ -0,0 +1 @@
+What is the recommended tag to use for the `hazy` repository?  Use the output from the recommendation function to answer this question.

prompts/recommended_tags/README.md

@@ -0,0 +1,29 @@
+---
+extractors:
+tool_choice: auto
+model: gpt-4
+stream: true
+functions:
+  - name: docker_scout_tag_recommendation
+---
+
+# Background
+
+```sh
+# docker:command=recommended-tags
+bb -m prompts run \
+              --host-dir /Users/slim/docker/labs-make-runbook \
+              --user jimclark106 \
+              --platform darwin \
+              --prompts-dir prompts/recommended_tags \
+              --url http://localhost:11434/v1/chat/completions \
+              --model "mistral-nemo" \
+              --nostream \
+```
+
+
+## functions
+
+These instructions in these prompts rely on 3 functions
+
+3. `docker_scout_tag_recommendation` - uses Docker Scout to improve the choice tag selection during code generation.