# io.github.moralito311-andr/andreax

54 pay-per-call AI services over x402 (USDC on Base): compression, OCR, FX, inference.

- **Type:** MCP server
- **Trust:** 60/100 (B), scored on the content rubric
- **Verification:** community-indexed — nobody has claimed this listing
- **Version:** 1.1.0
- **Author:** io.github.moralito311-andr
- **License:** Unknown
- **Endpoints:** streamable-http https://pagos.andreax.dev/mcp
- **Source:** https://pagos.andreax.dev/mcp
- **Endpoint health:** reachable (last checked 2026-09-26T21:35:31.601Z, 5 samples) — uptime is not a security property and is not part of the trust score
- **Compatible clients:** claude-code, cursor, copilot, chatgpt, gemini (basis: transport)

## Trust

60/100 (B), scored on the content rubric
- Publisher verified: no
- Install scripts: nothing suspicious found
- Prompt-injection scan: clean
- Obfuscation scan: clean
- Evidence age: 0 days

## Security scan

- **Status:** clean
- **Scanned:** 2026-09-26T21:35:31.601Z
- **Version scanned:** live
- **CVEs:** no coverage — this entry has no package coordinates to query OSV against, so "no known CVEs" is NOT asserted for it.

## Tools

200 declared. Observed from a live `tools/list` probe.
- `market-data` — Spot USDC/COP price and market depth (premium data). [x402: 0.001 USDC on Base, pay-per-use]
- `ai-inference` — Model answer to your query (pay-per-use AI compute). [x402: 0.002 USDC on Base, pay-per-use]
- `rwa-report` — Yield and risk of 12 tokenized RWA funds, updated. [x402: 0.005 USDC on Base, pay-per-use]
- `compress` — Compress a prompt into dense language and return the version with fewer tokens (typical 20-40% saving). Use it BEFORE sending prompts to any LLM to cut token co
- `distilla-compress` — Distilla (branded, real PyPI package): compress a prompt to dense language, same meaning, fewer tokens. input=prompt to compress. [x402: 0.02 USDC on Base, pay-
- `distilla-evocation-score` — Distilla: score how dense a prompt already is, 0-100 (higher = less filler to cut). input=prompt to score. [x402: 0.01 USDC on Base, pay-per-use]
- `distilla-identity` — Distilla: generate 7 dense identity aphorisms for a system from its description. input=system description. [x402: 0.05 USDC on Base, pay-per-use]
- `inference` — Return an LLM (Kimi/DeepSeek) answer to your query. General-purpose, pay-per-call inference for when an agent needs a model response. input=your query as text. 
- `fx` — Quote a currency conversion into USDC with spread (e.g. COP->USDC, USD->USDC). Get stablecoin exchange rates. input='<amount> <from> <to>'. [x402: 0.001 USDC on
- `research` — Research a topic on the web and return a synthesis with cited sources (titles + URLs). Pay-per-query research for when an agent needs current information with s
- `read-url` — Download a web page and return its clean, readable text (no HTML or scripts). Let an agent read a URL's content pay-per-call. input=http(s) URL. [x402: 0.002 US
- `report` — Generate a written, structured research report on a topic: researches the web and synthesizes a 400-600 word report with cited sources. Premium service for when
- `crypto-price` — Return the LIVE price of a cryptocurrency in USD and COP with its 24h change (real market data). High volume for agents that need prices. input=symbol or id (bt
- `embeddings` — Turn text into a vector (embedding) for semantic search, RAG, clustering or memory. The building block every AI agent needs. High volume. input=text. [x402: 0.0
- `semantic-search` — Rank a list of documents by MEANING closeness to a query (cosine similarity over embeddings). Ready-to-use RAG, no need for the agent to run its own vector DB. 
- `read-pdf` — Download a PDF from a URL and return its extracted text. Let an agent read documents (reports, contracts, papers) pay-per-call. input=http(s) URL of a PDF. [x40
- `wallet-balance` — Return the on-chain balance (ETH and USDC) of an address on the Base network. Live on-chain read for agents verifying funds or balances. input=0x address. [x402
- `ocr` — OCR: extract text from an image (JPG/PNG) or a PDF, including scanned PDFs with no text layer. Local engine (Tesseract 5), Spanish and English. Upload the file 
- `vision` — Computer vision: describe or analyze an image with a LOCAL multimodal model (llava). For agents that need to 'see' (describe scenes, read diagrams, classify ima
- `transcribe` — Transcribe an audio file to text (speech-to-text) with LOCAL Whisper, automatic language detection. For agents processing voice notes, calls or podcasts. Upload
- `smart-document` — Premium PIPELINE in a single call: takes an image or PDF, extracts its text (OCR), summarizes it and translates the summary to the target language. For agents t
- `meeting-minutes` — Premium PIPELINE in a single call: takes an audio (meeting, voice note, call), transcribes it with local Whisper and writes MINUTES with Summary, Decisions and 
- `deep-research-report` — PREMIUM: full research report with cited web sources. input=topic. [x402: 5.0 USDC on Base, pay-per-use]
- `screen-aware-assist` — PREMIUM: reads your screen (VLM) and assists. input=question. [x402: 2.0 USDC on Base, pay-per-use]
- `voice-clone-tts` — PREMIUM: returns SSML ready for your TTS engine. input=text. [x402: 3.0 USDC on Base, pay-per-use]
- `predictive-analytics` — PREMIUM: trend + short-term prediction + risks over your data. input=data/series. [x402: 10.0 USDC on Base, pay-per-use]
- `multi-agent-debate` — PREMIUM: multi-model debate (DeepSeek+Kimi) then a judged conclusion. input=question. [x402: 7.0 USDC on Base, pay-per-use]
- `code-review-security` — PREMIUM: security-focused code review by severity. input=code. [x402: 4.0 USDC on Base, pay-per-use]
- `market-intelligence` — PREMIUM: market intelligence with live data + scenarios. input=asset/market. [x402: 8.0 USDC on Base, pay-per-use]
- `autonomous-workflow` — PREMIUM: executable step-by-step autonomous workflow plan. input=goal. [x402: 15.0 USDC on Base, pay-per-use]
- `content-pipeline` — PREMIUM: full content pipeline (post + variants + hashtags). input=brief. [x402: 5.0 USDC on Base, pay-per-use]
- `compliance-audit` — PREMIUM: compliance/legal-risk audit by severity. input=document/policy. [x402: 20.0 USDC on Base, pay-per-use]
- `comprimir` — CONSUMER: compress a prompt to dense language keeping meaning (<=50%). input=text. B2B: save LLM tokens in pipelines. [x402: 0.001 USDC on Base, pay-per-use]
- `contar-tokens` — CONSUMER: exact token count (tiktoken) for a model. input=text, modelo=optional. B2B: budget LLM calls. [x402: 0.001 USDC on Base, pay-per-use]
- `interpretar` — CONSUMER: detect intent + categories + sentiment. input=message, categorias=optional. B2B: ticket routing, chatbots. [x402: 0.001 USDC on Base, pay-per-use]
- `fast-translate` — CONSUMER: translate + proofread to any language. input=text, to=lang. [x402: 0.05 USDC on Base, pay-per-use]
- `quick-summary` — CONSUMER: summary + keywords. input=text. [x402: 0.02 USDC on Base, pay-per-use]
- `email-polish` — CONSUMER: rewrite email formal + proofread. input=draft, tono=optional. [x402: 0.05 USDC on Base, pay-per-use]
- `doc-extract` — CONSUMER: extract structured JSON from a document. input=text, campos=optional. [x402: 0.1 USDC on Base, pay-per-use]
- `audio-to-text` — CONSUMER: clean + summarize a transcript. input/transcript=text. [x402: 0.2 USDC on Base, pay-per-use]
- …and 160 more

## Install

**Verdict: review** — Installable, but 2 things to check first: No CVE coverage: this entry has no npm/PyPI coordinates to query OSV against, so "no known vulnerabilities" is not a claim that can be made about it.
**Cautions** (coverage gaps and advisories — never blocking)
- No CVE coverage: this entry has no npm/PyPI coordinates to query OSV against, so "no known vulnerabilities" is not a claim that can be made about it.
- No publisher has proved control of this listing; it is indexed, not vouched for.
**Config** (claude-code):
```json
"{\n  \"mcpServers\": {\n    \"andreax\": {\n      \"type\": \"http\",\n      \"url\": \"https://pagos.andreax.dev/mcp\"\n    }\n  }\n}"
```

## Blast radius

Moderate to extensive — no credential declaration found, from the publisher, the upstream registry, or the README. Known so far: reaches a data store; runs on someone else's infrastructure.
- Floor 22, ceiling 40 (tier: unknown)
- `unknown` means the floor and ceiling land in different bands — not measured enough to name one. It does not mean low.
- This is impact, not likelihood. A high radius is not a defect: a filesystem server is supposed to write files. It is never part of the trust score.

## Machine-readable views of this entry

- Signed JSON: https://forgeregistry.com/api/v1/packages/io.github.moralito311-andr%2Fandreax
- Install plan: https://forgeregistry.com/api/v1/packages/io.github.moralito311-andr%2Fandreax/install-plan
- Alternatives: https://forgeregistry.com/api/v1/alternatives/io.github.moralito311-andr%2Fandreax
- HTML page: https://forgeregistry.com/registry/io.github.moralito311-andr%2Fandreax
- MCP: POST https://forgeregistry.com/api/mcp → `forge_get_package` / `forge_install_plan`

## About this document

Generated by Forge (https://forgeregistry.com) — a compact rendering of the same record served, signed, at the JSON URL above. Trust and scan facts are the registry's own measurements; anything Forge did not measure is named as unmeasured rather than omitted.