Add LiteLLM as an agent for model connections

.github/workflows/workflows/pr-agent.yaml

@@ -0,0 +1,9 @@
+name: AI PR Agent
+
+on:
+  pull_request:
+    types: [opened, reopened, ready_for_review]
+
+jobs:
+  pr_agent_job:
+    uses: SolaceDev/ai-build-actions/.github/workflows/ai_pr.yaml@use_sonnet_3_5

.pr_agent.toml

@@ -0,0 +1,134 @@
+[config]
+model="bedrock/anthropic.claude-3-5-sonnet-20240620-v1:0"
+model_turbo="bedrock/anthropic.claude-3-5-sonnet-20240620-v1:0"
+fallback_models="[bedrock/anthropic.claude-3-sonnet-20240229-v1:0]"
+git_provider="github"
+publish_output=true
+publish_output_progress=false
+verbosity_level=2 # 0,1,2
+use_extra_bad_extensions=false
+use_wiki_settings_file=true
+use_repo_settings_file=true
+use_global_settings_file=true
+ai_timeout=120 # 2minutes
+max_description_tokens = 800
+max_commits_tokens = 500
+max_model_tokens = 64000 # Limits the maximum number of tokens that can be used by any model, regardless of the model's default capabilities.
+patch_extra_lines = 200
+secret_provider="google_cloud_storage"
+cli_mode=false
+ai_disclaimer_title=""  # Pro feature, title for a collapsible disclaimer to AI outputs
+ai_disclaimer=""  # Pro feature, full text for the AI disclaimer
+
+[pr_reviewer] # /review #
+# enable/disable features
+require_score_review=false
+require_tests_review=true
+require_estimate_effort_to_review=true
+require_can_be_split_review=false
+# soc2
+require_soc2_ticket=false
+soc2_ticket_prompt="Does the PR description include a link to ticket in a project management system (e.g., Jira, Asana, Trello, etc.) ?"
+# general options
+num_code_suggestions=4
+inline_code_comments = true
+ask_and_reflect=false
+#automatic_review=true
+persistent_comment=true
+extra_instructions = ""
+final_update_message = true
+# review labels
+enable_review_labels_security=true
+enable_review_labels_effort=true
+# specific configurations for incremental review (/review -i)
+require_all_thresholds_for_incremental_review=false
+minimal_commits_for_incremental_review=0
+minimal_minutes_for_incremental_review=0
+enable_help_text=true # Determines whether to include help text in the PR review. Enabled by default.
+# auto approval
+enable_auto_approval=false
+maximal_review_effort=5
+
+[pr_description] # /describe #
+publish_labels=true
+add_original_user_description=true
+keep_original_user_title=true
+generate_ai_title=false
+use_bullet_points=true
+extra_instructions = ""
+enable_pr_type=true
+final_update_message = true
+enable_help_text=false
+enable_help_comment=false
+# describe as comment
+publish_description_as_comment=false
+publish_description_as_comment_persistent=true
+## changes walkthrough section
+enable_semantic_files_types=true
+collapsible_file_list='adaptive' # true, false, 'adaptive'
+inline_file_summary=false # false, true, 'table'
+# markers
+use_description_markers=false
+include_generated_by_header=true
+
+[pr_code_suggestions] # /improve #
+max_context_tokens=8000
+num_code_suggestions=4
+commitable_code_suggestions = false
+extra_instructions = ""
+rank_suggestions = false
+enable_help_text=true
+persistent_comment=false
+# params for '/improve --extended' mode
+auto_extended_mode=true
+num_code_suggestions_per_chunk=5
+max_number_of_calls = 3
+parallel_calls = true
+rank_extended_suggestions = false
+final_clip_factor = 0.8
+
+[pr_add_docs] # /add_docs #
+extra_instructions = ""
+docs_style = "Sphinx Style" # "Google Style with Args, Returns, Attributes...etc", "Numpy Style", "Sphinx Style", "PEP257", "reStructuredText"
+
+[pr_update_changelog] # /update_changelog #
+push_changelog_changes=false
+extra_instructions = ""
+
+[pr_analyze] # /analyze #
+
+[pr_test] # /test #
+extra_instructions = ""
+testing_framework = "" # specify the testing framework you want to use
+num_tests=3            # number of tests to generate. max 5.
+avoid_mocks=true       # if true, the generated tests will prefer to use real objects instead of mocks
+file = ""              # in case there are several components with the same name, you can specify the relevant file
+class_name = ""        # in case there are several methods with the same name in the same file, you can specify the relevant class name
+enable_help_text=true
+
+[pr_improve_component] # /improve_component #
+num_code_suggestions=4
+extra_instructions = ""
+file = ""              # in case there are several components with the same name, you can specify the relevant file
+class_name = ""
+
+[checks] # /checks (pro feature) #
+enable_auto_checks_feedback=true
+excluded_checks_list=["lint"] # list of checks to exclude, for example: ["check1", "check2"]
+persistent_comment=true
+enable_help_text=true
+
+[pr_help] # /help #
+
+[pr_config] # /config #
+
+[github]
+# The type of deployment to create. Valid values are 'app' or 'user'.
+deployment_type = "user"
+ratelimit_retries = 5
+base_url = "https://api.github.com"
+publish_inline_comments_fallback_with_verification = true
+try_fix_invalid_inline_comments = true
+
+[litellm]
+drop_params = true

docs/components/index.md

@@ -17,6 +17,8 @@
 | [langchain_vector_store_delete](langchain_vector_store_delete.md) | This component allows for entries in a LangChain Vector Store to be deleted. This is needed for the continued maintenance of the vector store. Due to the nature of langchain vector stores, you need to specify an embedding component even though it is not used in this component. |
 | [langchain_vector_store_embedding_index](langchain_vector_store_embedding_index.md) | Use LangChain Vector Stores to index text for later semantic searches. This will take text, run it through an embedding model and then store it in a vector database. |
 | [langchain_vector_store_embedding_search](langchain_vector_store_embedding_search.md) | Use LangChain Vector Stores to search a vector store with a semantic search. This will take text, run it through an embedding model with a query embedding and then find the closest matches in the store. |
+| [litellm_chat_model](litellm_chat_model.md) | LiteLLM chat model component |
+| [litellm_chat_model_with_history](litellm_chat_model_with_history.md) | LiteLLM model handler component with conversation history |
 | [message_filter](message_filter.md) | A filtering component. This will apply a user configurable expression. If the expression evaluates to True, the message will be passed on. If the expression evaluates to False, the message will be discarded. If the message is discarded, any previous components that require an acknowledgement will be acknowledged. |
 | [openai_chat_model](openai_chat_model.md) | OpenAI chat model component |
 | [openai_chat_model_with_history](openai_chat_model_with_history.md) | OpenAI chat model component with conversation history |
@@ -30,3 +32,5 @@
 | [websearch_bing](websearch_bing.md) | Perform a search query on Bing. |
 | [websearch_duckduckgo](websearch_duckduckgo.md) | Perform a search query on DuckDuckGo. |
 | [websearch_google](websearch_google.md) | Perform a search query on Google. |
+| [websocket_input](websocket_input.md) | Listen for incoming messages on a websocket connection. |
+| [websocket_output](websocket_output.md) | Send messages to a websocket connection. |

docs/components/litellm_chat_model.md

@@ -0,0 +1,84 @@
+# LiteLLMChatModel
+
+LiteLLM chat model component
+
+## Configuration Parameters
+
+```yaml
+component_name: <user-supplied-name>
+component_module: litellm_chat_model
+component_config:
+  action: <string>
+  load_balancer: <string>
+  embedding_params: <string>
+  temperature: <string>
+  stream_to_flow: <string>
+  stream_to_next_component: <string>
+  llm_mode: <string>
+  stream_batch_size: <string>
+  set_response_uuid_in_user_properties: <boolean>
+  history_max_turns: <string>
+  history_max_time: <string>
+  history_max_turns: <string>
+  history_max_time: <string>
+```
+
+| Parameter | Required | Default | Description |
+| --- | --- | --- | --- |
+| action | True | inference | The action to perform (e.g., 'inference', 'embedding') |
+| load_balancer | False |  | Add a list of models to load balancer. |
+| embedding_params | False |  | LiteLLM model parameters. The model, api_key and base_url are mandatory.find more models at https://docs.litellm.ai/docs/providersfind more parameters at https://docs.litellm.ai/docs/completion/input |
+| temperature | False | 0.7 | Sampling temperature to use |
+| stream_to_flow | False |  | Name the flow to stream the output to - this must be configured for llm_mode='stream'. This is mutually exclusive with stream_to_next_component. |
+| stream_to_next_component | False | False | Whether to stream the output to the next component in the flow. This is mutually exclusive with stream_to_flow. |
+| llm_mode | False | none | The mode for streaming results: 'sync' or 'stream'. 'stream' will just stream the results to the named flow. 'none' will wait for the full response. |
+| stream_batch_size | False | 15 | The minimum number of words in a single streaming result. Default: 15. |
+| set_response_uuid_in_user_properties | False | False | Whether to set the response_uuid in the user_properties of the input_message. This will allow other components to correlate streaming chunks with the full response. |
+| history_max_turns | False | 10 | Maximum number of conversation turns to keep in history |
+| history_max_time | False | 3600 | Maximum time to keep conversation history (in seconds) |
+| history_max_turns | False | 10 | Maximum number of conversation turns to keep in history |
+| history_max_time | False | 3600 | Maximum time to keep conversation history (in seconds) |
+
+
+## Component Input Schema
+
+```
+{
+  messages: [
+    {
+      role:       <string>,
+      content:       <string>
+    },
+    ...
+  ],
+  clear_history_but_keep_depth:   <integer>
+}
+```
+| Field | Required | Description |
+| --- | --- | --- |
+| messages | True |  |
+| messages[].role | True |  |
+| messages[].content | True |  |
+| clear_history_but_keep_depth | False | Clear history but keep the last N messages. If 0, clear all history. If not set, do not clear history. |
+
+
+## Component Output Schema
+
+```
+{
+  content:   <string>,
+  chunk:   <string>,
+  response_uuid:   <string>,
+  first_chunk:   <boolean>,
+  last_chunk:   <boolean>,
+  streaming:   <boolean>
+}
+```
+| Field | Required | Description |
+| --- | --- | --- |
+| content | True | The generated response from the model |
+| chunk | False | The current chunk of the response |
+| response_uuid | False | The UUID of the response |
+| first_chunk | False | Whether this is the first chunk of the response |
+| last_chunk | False | Whether this is the last chunk of the response |
+| streaming | False | Whether this is a streaming response |

docs/components/litellm_chat_model_with_history.md

@@ -0,0 +1,84 @@
+# LiteLLMChatModelWithHistory
+
+LiteLLM model handler component with conversation history
+
+## Configuration Parameters
+
+```yaml
+component_name: <user-supplied-name>
+component_module: litellm_chat_model_with_history
+component_config:
+  action: <string>
+  load_balancer: <string>
+  embedding_params: <string>
+  temperature: <string>
+  stream_to_flow: <string>
+  stream_to_next_component: <string>
+  llm_mode: <string>
+  stream_batch_size: <string>
+  set_response_uuid_in_user_properties: <boolean>
+  history_max_turns: <string>
+  history_max_time: <string>
+  history_max_turns: <string>
+  history_max_time: <string>
+```
+
+| Parameter | Required | Default | Description |
+| --- | --- | --- | --- |
+| action | True | inference | The action to perform (e.g., 'inference', 'embedding') |
+| load_balancer | False |  | Add a list of models to load balancer. |
+| embedding_params | False |  | LiteLLM model parameters. The model, api_key and base_url are mandatory.find more models at https://docs.litellm.ai/docs/providersfind more parameters at https://docs.litellm.ai/docs/completion/input |
+| temperature | False | 0.7 | Sampling temperature to use |
+| stream_to_flow | False |  | Name the flow to stream the output to - this must be configured for llm_mode='stream'. This is mutually exclusive with stream_to_next_component. |
+| stream_to_next_component | False | False | Whether to stream the output to the next component in the flow. This is mutually exclusive with stream_to_flow. |
+| llm_mode | False | none | The mode for streaming results: 'sync' or 'stream'. 'stream' will just stream the results to the named flow. 'none' will wait for the full response. |
+| stream_batch_size | False | 15 | The minimum number of words in a single streaming result. Default: 15. |
+| set_response_uuid_in_user_properties | False | False | Whether to set the response_uuid in the user_properties of the input_message. This will allow other components to correlate streaming chunks with the full response. |
+| history_max_turns | False | 10 | Maximum number of conversation turns to keep in history |
+| history_max_time | False | 3600 | Maximum time to keep conversation history (in seconds) |
+| history_max_turns | False | 10 | Maximum number of conversation turns to keep in history |
+| history_max_time | False | 3600 | Maximum time to keep conversation history (in seconds) |
+
+
+## Component Input Schema
+
+```
+{
+  messages: [
+    {
+      role:       <string>,
+      content:       <string>
+    },
+    ...
+  ],
+  clear_history_but_keep_depth:   <integer>
+}
+```
+| Field | Required | Description |
+| --- | --- | --- |
+| messages | True |  |
+| messages[].role | True |  |
+| messages[].content | True |  |
+| clear_history_but_keep_depth | False | Clear history but keep the last N messages. If 0, clear all history. If not set, do not clear history. |
+
+
+## Component Output Schema
+
+```
+{
+  content:   <string>,
+  chunk:   <string>,
+  response_uuid:   <string>,
+  first_chunk:   <boolean>,
+  last_chunk:   <boolean>,
+  streaming:   <boolean>
+}
+```
+| Field | Required | Description |
+| --- | --- | --- |
+| content | True | The generated response from the model |
+| chunk | False | The current chunk of the response |
+| response_uuid | False | The UUID of the response |
+| first_chunk | False | Whether this is the first chunk of the response |
+| last_chunk | False | Whether this is the last chunk of the response |
+| streaming | False | Whether this is a streaming response |

docs/components/openai_chat_model.md

@@ -56,9 +56,19 @@ component_config:
 
 ```
 {
-  content:   <string>
+  content:   <string>,
+  chunk:   <string>,
+  response_uuid:   <string>,
+  first_chunk:   <boolean>,
+  last_chunk:   <boolean>,
+  streaming:   <boolean>
 }
 ```
 | Field | Required | Description |
 | --- | --- | --- |
 | content | True | The generated response from the model |
+| chunk | False | The current chunk of the response |
+| response_uuid | False | The UUID of the response |
+| first_chunk | False | Whether this is the first chunk of the response |
+| last_chunk | False | Whether this is the last chunk of the response |
+| streaming | False | Whether this is a streaming response |

docs/components/openai_chat_model_with_history.md

@@ -62,9 +62,19 @@ component_config:
 
 ```
 {
-  content:   <string>
+  content:   <string>,
+  chunk:   <string>,
+  response_uuid:   <string>,
+  first_chunk:   <boolean>,
+  last_chunk:   <boolean>,
+  streaming:   <boolean>
 }
 ```
 | Field | Required | Description |
 | --- | --- | --- |
 | content | True | The generated response from the model |
+| chunk | False | The current chunk of the response |
+| response_uuid | False | The UUID of the response |
+| first_chunk | False | Whether this is the first chunk of the response |
+| last_chunk | False | Whether this is the last chunk of the response |
+| streaming | False | Whether this is a streaming response |