File size: 2,014 Bytes
e8edb00
 
 
 
f481f4b
e8edb00
 
08f932d
e8edb00
 
 
 
 
 
 
 
 
 
 
 
 
08f932d
 
 
 
 
 
 
 
 
 
 
 
 
 
 
e8edb00
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
components:
  llm:
    init_parameters:
      api_params:
        model: meta-llama/Llama-4-Maverick-17B-128E-Instruct
      api_type: serverless_inference_api
      generation_kwargs:
        max_new_tokens: 5000
        stop:
        - <stop>
        stop_sequences: []
      streaming_callback: null
      token:
        env_vars:
        - HF_API_TOKEN
        strict: true
        type: env_var
    type: haystack.components.generators.hugging_face_api.HuggingFaceAPIGenerator
  prompt_builder:
    init_parameters:
      required_variables: null
      template: "\nYou are an instructional designer tasked with creating engaging\
        \ and comprehensive content for training modules. \nYour goal is to generate\
        \ detailed educational material that fully explains each sub module, ensuring\
        \ that employees gain \na complete understanding of the subject matter.\n\n\
        Task: Produce valid Markdown output containing the **content** for the following\
        \ training module.\n\nModule Details:\n{{module}}\n\nInstructions:\n- Return\
        \ only a Markdown document that follows the provided format (no extra text).\n\
        - Do not include code fences or additional commentary.\n- Ensure that the\
        \ content in each sub module is sufficiently detailed to properly educate\
        \ an employee.\n\nExample Format:\n<start>\n## Content\n\n### Subtopic 1\n\
        Detailed content explaining the topic comprehensively.\n\n### Subtopic 2\n\
        In-depth explanation covering all necessary details.\n<stop>\n\n{% if invalid_replies\
        \ and error_message %}\nYour previous Markdown output was invalid. Error:\
        \ {{error_message}}. \nPlease return only valid Markdown, with no extra text,\
        \ enclosed by <start> and <stop>.\n{% endif %}\n"
      variables: null
    type: haystack.components.builders.prompt_builder.PromptBuilder
connections:
- receiver: llm.prompt
  sender: prompt_builder.prompt
max_runs_per_component: 100
metadata: {}