51 Services from pagos.andreax.dev

Fetches a web page by URL and returns its clean, readable text content stripped of HTML, scripts, and markup

Generates a structured 400-600 word written research report on any topic by searching the web and synthesizing findings with cited sources.

Ranks a list of documents by semantic similarity to a query using cosine similarity over embeddings, enabling ready-to-use RAG without maintaining a vector database.

Transcribes an audio file to text using a locally-run Whisper model, with automatic language detection and optional language hint.

Converts a text string into a numerical embedding vector for use in semantic search, RAG pipelines, clustering, and AI memory.

Sends a text prompt to a Kimi or DeepSeek LLM and returns the model's generated response, billed per call via x402.

Rewrites a given text in formal and professional tone while preserving the original language

Generates a regular expression pattern from a natural language description of what it should capture

Accepts an image or PDF, extracts its text via OCR, summarizes it, and translates the summary into a target language — all in a single API call.

Converts a natural language frequency description into a valid cron expression

Estimates the token count of a text string to help budget LLM API calls before making them

Translates a long document into any target language using a premium LLM-powered translation service.

Performs a structured risk analysis of a business decision or plan, returning identified risks with mitigation strategies.

Detects the ISO language code of a given text string, optimized for high-volume agent use at low cost.

Generates FAQ entries (questions and answers) from a given text or topic, useful for documentation and support content.

Extracts named entities (NER) from a given text and returns them as structured JSON

Extracts 5-10 SEO/indexing keywords from a given text using NLP

Classifies the sentiment of a given text as positive, negative, or neutral with a confidence score.

Returns the live on-chain ETH and USDC balances for a given EVM wallet address on the Base network

Returns the live price of a cryptocurrency in USD and COP along with its 24-hour change, sourced from real market data.

Generates a competitive analysis report for a given industry sector or company, including competitive positioning and recommendations.

Generates a professional investor pitch or one-pager document from a project or startup description.

Researches a topic on the web and returns a synthesized summary with cited sources (titles and URLs)

Explains any given concept in simple, plain language using an LLM, optimized for high call volume.

Extracts text from images (JPG/PNG) or PDFs—including scanned PDFs without a text layer—using a local Tesseract 5 engine supporting Spanish and English.

Analyzes or describes the content of an image using a local multimodal LLM (LLaVA), answering a user-specified question about the image

Transcribes an audio recording with Whisper and generates a structured meeting minutes document (Resumen, Decisiones, Tareas) in a single call

Generates a structured strategic plan with phases, actions, and metrics from a given objective or context.

Summarizes any input text into clear, faithful bullet points for condensing documents or threads

Repairs and validates broken JSON, returning a clean, well-formed JSON object

Returns a spread-adjusted currency conversion quote from a source currency (e.g. COP, USD) to USDC stablecoin

Generates a complete sales/prospecting email (subject line + body) from a given context using an LLM.

Downloads a PDF from a given URL and returns its extracted plain text content

Converts a natural language request into a SQL query using an LLM backend

Translates a natural language instruction into a dense, actionable structured contract or data structure

Translates any input text into fluent, natural English using a language model.

Generates a professional contract or agreement draft in Spanish from a specified contract type and key terms, with legal disclaimer

Parses free-form text into structured JSON by extracting key data fields using an LLM-backed extraction pipeline

Verifies a factual claim and returns a verdict with a supporting reason and confidence level.

Corrects spelling and grammar errors in a given text using an LLM, returning the corrected version.

Generates source code from a natural-language description of what the code should do

Analyzes text input and flags unsafe or inappropriate content for agent-based filtering pipelines.

Compresses a text prompt into a denser, lower-token equivalent, typically achieving 20-40% token savings without significant loss of meaning.

Generates a professional draft reply to a customer support message (email, chat, or ticket) given the original message text.

Translates text into any target language using a simple 'target_language | text' input format

Translates or interprets legal/juridical text into plain language or vice versa using an LLM-based service.

Classifies the intent of a natural language message and extracts structured parameters from it.

Reviews submitted source code for bugs, security vulnerabilities, and improvement suggestions using LLM-powered analysis.

Generates 5 attractive and honest titles for a given topic or text content

Extracts the irreducible conceptual essence (core principle) of any given concept or idea

Classifies a text string into a topic or category for routing, labeling, or filtering at high volume.