Struct ContentGenerator 

pub struct ContentGenerator { /* private fields */ }

Content generator that combines an LLM provider with business context.

Implementations

impl ContentGenerator

pub fn new(provider: Box<dyn LlmProvider>, business: BusinessProfile) -> Self

Create a new content generator.

pub fn business(&self) -> &BusinessProfile

Returns a reference to the business profile.

pub async fn generate_reply( &self, tweet_text: &str, tweet_author: &str, mention_product: bool, ) -> Result<GenerationOutput, LlmError>

Generate a reply to a tweet.

pub async fn generate_reply_with_archetype( &self, tweet_text: &str, tweet_author: &str, mention_product: bool, archetype: Option<ReplyArchetype>, ) -> Result<GenerationOutput, LlmError>

Generate a reply using a specific archetype for varied output.

pub async fn generate_reply_with_context( &self, tweet_text: &str, tweet_author: &str, mention_product: bool, archetype: Option<ReplyArchetype>, rag_context: Option<&str>, ) -> Result<GenerationOutput, LlmError>

Generate a reply with optional RAG context injected into the prompt.

pub async fn generate_tweet( &self, topic: &str, ) -> Result<GenerationOutput, LlmError>

Generate a standalone educational tweet.

pub async fn generate_tweet_with_format( &self, topic: &str, format: Option<TweetFormat>, ) -> Result<GenerationOutput, LlmError>

Generate a tweet using a specific format for varied structure.

pub async fn generate_tweet_with_context( &self, topic: &str, format: Option<TweetFormat>, rag_context: Option<&str>, ) -> Result<GenerationOutput, LlmError>

Generate a tweet with optional RAG context injected into the prompt.

pub async fn improve_draft( &self, draft: &str, tone_cue: Option<&str>, ) -> Result<GenerationOutput, LlmError>

Rewrite/improve an existing draft tweet with an optional tone cue.

pub async fn improve_draft_with_context( &self, draft: &str, tone_cue: Option<&str>, rag_context: Option<&str>, ) -> Result<GenerationOutput, LlmError>

Rewrite/improve an existing draft tweet with optional RAG context injected into the system prompt.

pub async fn generate_hooks( &self, topic: &str, rag_context: Option<&str>, ) -> Result<HookGenerationOutput, LlmError>

Generate 5 differentiated hook options for the given topic.

pub async fn generate_mined_angles( &self, topic: &str, neighbors: &[NeighborContent], selection_context: Option<&str>, ) -> Result<AngleMiningOutput, LlmError>

Generate evidence-backed content angles from neighbor notes.

pub async fn extract_highlights( &self, rag_context: &str, ) -> Result<Vec<String>, LlmError>

Extract 3-5 concise, tweetable key highlights from provided context.

Used as an intermediate curation step before generating content from vault notes, letting the user review and select which insights to include.

pub async fn generate_thread( &self, topic: &str, ) -> Result<ThreadGenerationOutput, LlmError>

Generate an educational thread of 5-8 tweets.

pub async fn generate_thread_with_structure( &self, topic: &str, structure: Option<ThreadStructure>, ) -> Result<ThreadGenerationOutput, LlmError>

Generate a thread using a specific structure for varied content.

pub async fn generate_thread_with_context( &self, topic: &str, structure: Option<ThreadStructure>, rag_context: Option<&str>, ) -> Result<ThreadGenerationOutput, LlmError>

Generate a thread with optional RAG context injected into the prompt.

pub async fn generate_thread_with_hook( &self, topic: &str, opening_hook: &str, structure: Option<ThreadStructure>, rag_context: Option<&str>, ) -> Result<ThreadGenerationOutput, LlmError>

Generate a thread that starts with a specific opening hook tweet.

The hook is used verbatim as the first tweet, and the LLM generates 4-7 additional tweets continuing from that opening.