Textcortex connects to a Nagent workspace with an API key. Once it is connected, agents can call 7 Textcortex actions, such as "Create Product Title", "Create Social Media Post" and "Create Video Description". Nothing is enabled on connect: each action is allowed one at a time, and an action with side effects runs or waits for a person according to the agent's level.
Every operation an agent can call against Textcortex, with input parameters and output schema.
TEXTCORTEX_CREATE_PRODUCT_TITLETool to generate catchy product titles for listings. Use when you need to create compelling titles for product listings or descriptions.
Input parameters
The number of outputs to generate (default: 1).
Name of the product.
Brand of the product.
Model specification to use for generation.
Category of the product.
Keywords to include in the product description.
Tone/style preference for the generated title.
The maximum number of tokens to generate (default: 2048).
Features of the product.
Source language (defaults to English). ISO language code.
Target language for output. ISO language code.
Controls output randomness (0.0 to 1.0).
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
TEXTCORTEX_CREATE_SOCIAL_MEDIA_POSTTool to generate engaging social media posts for Twitter and LinkedIn. Use when you need to create platform-specific content with appropriate tone, length, and formatting.
Input parameters
Number of different post variations to generate (default: 1). Useful when you want multiple options to choose from.
Target social media platform for the post. Determines the style, tone, and format of the generated content.
Model identifier to use for generation (e.g., 'gpt-4o'). If not specified, uses the default model.
The context or main idea for the social media post. Provide details about what the post should be about, key messages, or any specific information to include.
List of keywords or hashtags to include in the post. These will be naturally integrated into the generated content.
Tone or formality level for the post (e.g., 'formal', 'casual', 'professional'). Influences the writing style and language used.
Maximum number of tokens to generate (default: 2048). Controls the length of the output.
Language of the source context (ISO code, e.g., 'en', 'es', 'fr'). Default is 'en'.
Target language for the generated post (ISO code). If different from source_lang, the post will be translated.
Sampling temperature for output variation (0.0 to 1.0). Higher values produce more creative/random outputs, lower values are more deterministic.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
TEXTCORTEX_CREATE_VIDEO_DESCRIPTIONTool to generate compelling video descriptions for YouTube or other platforms. Use when you need to create optimized descriptions that include context, keywords, and target audience information.
Input parameters
The number of video description outputs to generate (default: 1).
Video platform mode for description generation.
The model identifier to use for generation.
The title of the video for which to generate the description.
Topic, target audience, and other details about the video to help generate an appropriate description.
List of keywords to include in the video description for SEO purposes.
The formality level or tone for the generated description (e.g., 'formal', 'casual', 'professional').
The maximum number of tokens to generate (default: 2048).
The language of the source context and title (ISO code, e.g., 'en', 'auto').
The desired output language for the description (ISO code, e.g., 'en', 'es', 'fr').
The sampling temperature for text generation (0.0 to 1.0). Higher values make output more creative.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
TEXTCORTEX_GENERATE_CODETool to generate code for a given programming language. Use when you need to create code snippets, functions, or algorithms in specific programming languages.
Input parameters
Number of code completions to generate (default: 1).
The programming language to generate code in (e.g., python, javascript, java, sql, go, rust, php, typescript, c++, c#).
The code description or prompt to generate code from. Describe the functionality you want to implement.
The model to use for code generation (e.g., icortex-1, sophos-1).
Maximum number of tokens to generate in the response.
Sampling temperature for generation (0.0 to 1.0). Higher values make output more random.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
TEXTCORTEX_GENERATE_EMAILTool to generate email body and subject based on provided details. Use when you need to compose or personalize an email from a prompt.
Input parameters
Main prompt or context for generating the email. Include details like recipient, purpose, tone preference, and any specific information to include.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
TEXTCORTEX_SUMMARIZE_TEXTTool to summarize text. Use when you need a concise summary of longer content. Use after confirming the full input text.
Input parameters
The number of outputs to generate (default: 1).
Mode for summarization: 'default' or 'embeddings'.
The text to summarize. Either text or file_id must be provided.
Model identifier to use for summarization.
ID of the file to summarize. Either text or file_id must be provided.
Tone/style parameter for the summary.
The maximum number of tokens to generate (default: 2048).
The language of the source text (ISO code, e.g., 'en', 'auto').
The desired output language (ISO code, e.g., 'en', 'es', 'fr').
The sampling temperature for text generation (0.0 to 1.0).
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
TEXTCORTEX_TRANSLATE_TEXTTool to translate text into another language. Use when you need to translate content between different languages. Supports 72+ languages with auto-detection capability.
Input parameters
Number of translation variations to generate (default: 1).
The text to translate.
Model identifier to use for translation.
Maximum number of tokens to generate.
Source language code (ISO code, e.g., 'en', 'es'). Auto-detected if not provided.
Target language code to translate to (e.g., 'es' for Spanish, 'fr' for French, 'de' for German).
Sampling temperature for generation (0.0 to 1.0).
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
Agents can call 7 Textcortex actions on Nagent, including "Create Product Title", "Create Social Media Post" and "Create Video Description". Create Product Title: Generate catchy product titles for listings. Each action is listed on this page with its input parameters and its output.
Textcortex connects with an API key, under your workspace's own connection. Nothing is enabled on connect: each action is allowed one at a time and can be scoped to the agents that need it.
It also takes 12 optional inputs: n, name, brand, model, category, keywords, formality, max_tokens, description, source_lang, target_lang and temperature. It returns data, error and successful.