All services return structured TOON or JSON with full cost and latency metadata.
Use POST /process to run a service. Use POST /quote to estimate cost first.
url.readability.extract
URL Readability Extract
active
$0.003 / request · USDC
Fetch a public URL and extract clean main text, title, excerpt, byline, word count, links, and metadata using Mozilla Readability. Validates and blocks unsafe URLs (localhost, private IPs, metadata endpoints).
{ url: string } → output: text+metadata
readability extraction web text
document.basic_extract
Document Basic Extract
active
$0.002 / request · USDC
Extract basic structured data from plain text using deterministic regex rules (no LLM/AI). Finds emails, URLs, dates, amounts, phone numbers, hashtags, mentions, headings, and word/line/char counts.
{ text: string } → output: structured-fields
extraction regex structured-data text
web.metadata.audit
Web Metadata Audit
active
$0.003 / request · USDC
Fetch a public HTML page and extract title, meta description, canonical, OpenGraph, Twitter/X card tags, JSON-LD blocks and @types, robots meta, language, heading/link/image counts, metadata quality score, and missing_fields. Same URL safety as url.readability.extract.
{ url: string, include_links?: boolean, max_chars?: number } → output: metadata+json
web metadata seo deterministic phase-3-6
robots.policy.check
Robots.txt Policy Check
active
$0.002 / request · USDC
Fetch /robots.txt for the URL origin (safe http/https only), parse Allow/Disallow groups, evaluate whether the target URL path is allowed for a user-agent (default TechForAGI-Agent). Missing robots.txt → allowed with decision allow_by_default.
{ url: string, user_agent?: string } → output: policy+json
robots compliance crawler phase-3-6
captions.parse
Captions Parse
active
$0.002 / request · USDC
Deterministically parse SRT, WebVTT, or plain transcript text into segments (start, end, text), counts, and duration when timestamps exist. No external fetch; no YouTube.
{ text: string, format_hint?: "srt"|"webvtt"|"plain"|"auto", language?: string } → output: segments+json
captions srt webvtt transcript phase-3-6
youtube.captions.get
YouTube Captions (Deprecated)
deprecated
$0 / request · USDC
DEPRECATED: YouTube caption retrieval is no longer available. POST /process returns HTTP 410 for this service_id.
{ url: string, lang?: string } → output: n/a
deprecated youtube
youtube.video.formats
YouTube Video Formats (Deprecated)
deprecated
$0 / request · USDC
DEPRECATED: YouTube format listing is no longer available. POST /process returns HTTP 410 for this service_id.
{ url: string, quality_filter?: string } → output: n/a
deprecated youtube
text.summarize
Text Summarize (Deprecated)
deprecated
$0.001 / request · USDC
DEPRECATED: Mock summarization service kept for backwards compatibility. Use document.basic_extract for real extraction.
text → output: summary
deprecated mock
doc.extract
Document Extract (Deprecated)
deprecated
$0.004 / request · USDC
DEPRECATED: Mock extraction service kept for backwards compatibility. Use document.basic_extract for real extraction.
text → output: json/toon
deprecated mock
knowledge.search
Knowledge Search (Deprecated)
deprecated
$0.002 / request · USDC
DEPRECATED: Mock search service kept for backwards compatibility.
query → output: facts+sources
deprecated mock
# JSON format
GET https://techforagi.com/services.json
Accept: application/json# TOON format (compact, ~40% fewer tokens)
GET https://techforagi.com/services.toon
Accept: text/toon# TOON response example
services[3]{id,name,description,input,output,price_usd,status}:
text.summarize,Text Summarize,Summarize text...,text,summary,0.001,active
doc.extract,Document Extract,Extract fields...,text,json/toon,0.004,active
knowledge.search,Knowledge Search,Search sources...,query,facts+sources,0.002,active