New Runtime public records
AI-engineering archive: search, what is new, forecasts, URL context; paid x402 research tools.
사용해야 할까요
품질 및 안전성
발견 사항 (10)
- HIGH
- MEDIUMsearch_archive에서
- MEDIUMget_record에서
- MEDIUMget_topic_brief에서
- MEDIUMget_week_corpus에서
- MEDIUMsubmit_source에서
- MEDIUMsubmit_correction에서
- MEDIUMsubmit_usage_receipt에서
- MEDIUMsubmit_relevance_report에서
- LOWcontextualize_url에서
도구 정의와 프로토콜 준수에 대한 자동 분석을 기반으로 합니다.
컨텍스트 비용
이는 서버의 도구가 모델의 컨텍스트에 로드될 때마다 소비되는 대략적인 토큰 수입니다. 수치가 높을수록 다른 작업에 사용할 수 있는 주의가 줄어듭니다.
설치
원클릭 설치
`claude_desktop_config.json` 파일에 다음을 추가하세요:
{
"mcpServers": {
"archive": {
"url": "https://newruntime.com/mcp"
}
}
}원격 엔드포인트
https://newruntime.com/mcpstreamable-http할 수 있는 일
도구 목록
도구 (16)
🟢search_archive(query, kind, limit, offset, since)
Use when you need dated, source-linked coverage of an AI-engineering product, company or topic (agents, MCP, coding tools, model releases) without crawling the web. Keyword search (English, word forms match) over Field Notes (every news story), Signals, analyses, patterns, shifts, dossiers, topic hubs, tool change logs and tested-tool experiments; title and phrase matches rank first. Up to 25 hits per call with title, date, summary and URL; pass next_offset as offset for more. Same results as GET https://newruntime.com/search.json. Free, no auth. Example: {"query": "MCP security", "kind": "post", "limit": 5}. Then get_record for the full text.
입력 스키마
{
"type": "object",
"properties": {
"query": {
"type": "string",
"minLength": 2,
"maxLength": 200,
"description": "Words to look for (product, company, topic).",
"examples": [
"MCP security",
"Claude Code plugins",
"agent memory"
]
},
"kind": {
"type": "string",
"enum": [
"any",
"post",
"signal",
"analysis",
"pattern",
"shift",
"system",
"project",
"topic",
"tool",
"tested"
],
"default": "any",
"description": "post = Field Note (one per news story), signal = synthesis of 2+ notes, analysis = deep dive, pattern/shift = trends, system/project = dossiers, topic = hub."
},
"limit": {
"type": "integer",
"minimum": 1,
"maximum": 25,
"default": 5
},
"offset": {
"type": "integer",
"minimum": 0,
"maximum": 1000,
"default": 0,
"description": "Skip this many hits: the next_offset of the previous call."
},
"since": {
"type": "string",
"maxLength": 30,
"description": "Only records dated on or after this ISO date.",
"examples": [
"2026-10-01"
]
}
},
"required": [
"query"
],
"additionalProperties": false
}출력 스키마
{
"type": "object",
"properties": {
"query": {
"type": "string",
"description": "The query as searched."
},
"total_matches": {
"type": "number",
"description": "Records matching the query (all pages)."
},
"offset": {
"type": "number",
"description": "Hits skipped before this page."
},
"count": {
"type": "number",
"description": "Hits on this page."
},
"next_offset": {
"description": "Pass as offset for the next page; null on the last page."
},
"results": {
"type": "array",
"description": "Up to `limit` hits, best first.",
"items": {
"type": "object",
"properties": {
"title": {
"description": "Record title."
},
"url": {
"description": "newruntime.com page; pass it to get_record."
},
"kind": {
"description": "Record kind."
},
"published_at": {
"description": "Record date (same as date; GET /search.json name)."
},
"summary": {
"description": "One-line summary (same as description; GET /search.json name)."
},
"score": {
"description": "Relevance score; higher is better, comparable within one query."
},
"date": {
"description": "Record date."
},
"description": {
"description": "One-line summary."
},
"slug": {
"description": "Record slug or id."
},
"json": {
"description": "Machine-readable twin of the page."
}
}
}
},
"hint": {
"type": "string",
"description": "When nothing matched: what to try."
}
},
"required": [
"query",
"results"
]
}🟢get_record(url, kind, slug, level)
Use after search_archive, whats_new, what_changed, get_topic_brief or contextualize_url to read one record in full: metadata (dates, topics, source URLs, primary_source_url) and the body as Markdown (capped; see truncated). Pass any newruntime.com record URL they return (notes, signals, analyses, patterns, shifts, systems, projects, topics, tools, guides, sources), or kind + slug. level "short" returns only the title, dates, summary, topics, first source URLs, primary source and routes (2-3 KB instead of up to 60 KB; no body): use it to triage several records, then read the ones you keep at the default level "full". Free, no auth. Example: {"url": "https://newruntime.com/patterns/company-memory-needs-write-loops/", "level": "short"}.
입력 스키마
{
"type": "object",
"properties": {
"url": {
"type": "string",
"maxLength": 500,
"description": "A https://newruntime.com/<section>/<slug>/ record URL.",
"examples": [
"https://newruntime.com/analysis/claude-plugins-now-depend-on-anthropic-s-distribution-rules/"
]
},
"kind": {
"type": "string",
"enum": [
"post",
"signal",
"analysis",
"pattern",
"shift",
"system",
"project",
"topic",
"delta",
"graveyard",
"tool",
"learn",
"week",
"source",
"company"
]
},
"slug": {
"type": "string",
"maxLength": 200
},
"level": {
"type": "string",
"enum": [
"full",
"short"
],
"default": "full",
"description": "full (default): metadata and the body as Markdown. short: title, dates, summary, topics, first sources and routes only."
}
},
"additionalProperties": false
}출력 스키마
{
"type": "object",
"properties": {
"kind": {
"type": "string",
"description": "Record kind."
},
"url": {
"type": "string",
"description": "Canonical newruntime.com URL."
},
"level": {
"type": "string",
"description": "short when level short was asked; absent for the full record."
},
"record": {
"type": "object",
"description": "Metadata: title, dates, source_urls, topics, routes (level short: scalar fields, topics, first source URLs and routes only).",
"properties": {}
},
"summary": {
"description": "Level short: one-paragraph summary, or null."
},
"source_urls_total": {
"type": "number",
"description": "Level short: how many source URLs the record cites when more than the five listed."
},
"body_chars": {
"type": "number",
"description": "Length of the full body in characters."
},
"json_url": {
"description": "Machine-readable twin of the record, or null."
},
"next": {
"type": "string",
"description": "Level short: how to read the body."
},
"primary_source_url": {
"description": "First source that is not press, newsletter or social coverage; null when the record cites none."
},
"primary_source_missing": {
"type": "boolean",
"description": "True when every source of the record is secondary coverage."
},
"body_markdown": {
"description": "Body text as Markdown, or null."
},
"truncated": {
"type": "boolean",
"description": "True when body_markdown was cut; markdown_url has it all."
},
"markdown_url": {
"type": "string",
"description": "Full Markdown twin."
}
},
"required": [
"kind",
"url",
"record"
]
}🟢list_paid_tools
Use when the free tools left a question open and you can pay per call: lists New Runtime's x402 tools (find the primary source of a claim $0.10, rumour check $0.40, research bundle, prior-coverage check) with endpoint, USDC price, input example, when you are charged, and how to pay without an account. Free to call.
입력 스키마
{
"type": "object",
"properties": {},
"additionalProperties": false
}출력 스키마
{
"type": "object",
"properties": {
"how_to_pay": {
"type": "string",
"description": "How to pay per call over x402 without an account."
},
"tools": {
"type": "array",
"description": "Paid tools.",
"items": {
"type": "object",
"properties": {
"name": {
"description": "Tool name."
},
"endpoint": {
"description": "HTTPS endpoint; POST a JSON body."
},
"price_usd": {
"description": "Price in USD (USDC)."
},
"input_example": {
"description": "Example request body."
},
"returns": {
"description": "What the tool returns."
},
"charged": {
"description": "When you are charged."
}
}
}
}
},
"required": [
"how_to_pay",
"tools"
]
}🟢contextualize_url(url, task)
Use first when you found an AI or software URL elsewhere (a release note, repo, blog post, tweet) and want to know what New Runtime already has on it: records citing that exact URL or the same site, related records and patterns, the source hub, and a coverage summary. It reports what the archive holds, not a duplicate-or-new verdict (that is an optional paid call). Free, no auth. Example: {"url": "https://github.com/modelcontextprotocol/servers", "task": "is this new?"}.
입력 스키마
{
"type": "object",
"properties": {
"url": {
"type": "string",
"maxLength": 2000,
"description": "The https:// URL you are researching.",
"examples": [
"https://github.com/modelcontextprotocol/servers"
]
},
"task": {
"type": "string",
"maxLength": 300,
"description": "Optional: what you are trying to find out; its words refine related records.",
"examples": [
"is this new?",
"who else covered this"
]
}
},
"required": [
"url"
],
"additionalProperties": false
}출력 스키마
{
"type": "object",
"properties": {
"schema_version": {
"description": "Response schema version."
},
"url": {
"type": "string",
"description": "The URL as given."
},
"canonical_url": {
"type": "string",
"description": "The URL without tracking parameters and fragment."
},
"origin": {
"type": "string",
"description": "Host of the URL."
},
"task": {
"type": "string",
"description": "The task as given, when one was given."
},
"coverage": {
"type": "object",
"description": "status cited | same_site | related_only | not_found, counts, summary.",
"properties": {}
},
"source_hub": {
"description": "New Runtime's hub for this source host, or null."
},
"exact_records": {
"type": "array",
"description": "Published records whose source list contains this URL.",
"items": {
"type": "object",
"properties": {
"title": {
"description": "Record title."
},
"date": {
"description": "Record date, YYYY-MM-DD."
},
"url": {
"description": "newruntime.com page; pass it to get_record."
},
"json": {
"description": "Machine-readable twin of the page."
},
"why": {
"description": "One line on why the item matters."
}
}
}
},
"same_site_records": {
"type": "array",
"description": "Records citing other pages of the same site."
},
"related_records": {
"type": "array",
"description": "Keyword matches on the URL path and task.",
"items": {
"type": "object",
"properties": {
"title": {
"description": "Record title."
},
"date": {
"description": "Record date, YYYY-MM-DD."
},
"url": {
"description": "newruntime.com page; pass it to get_record."
},
"json": {
"description": "Machine-readable twin of the page."
},
"why": {
"description": "One line on why the item matters."
}
}
}
},
"related_patterns": {
"type": "array",
"description": "Patterns linked to the matched records.",
"items": {
"type": "object",
"properties": {
"title": {
"description": "Record title."
},
"date": {
"description": "Record date, YYYY-MM-DD."
},
"url": {
"description": "newruntime.com page; pass it to get_record."
},
"json": {
"description": "Machine-readable twin of the page."
},
"why": {
"description": "One line on why the item matters."
}
}
}
},
"query_terms": {
"type": "array",
"items": {
"type": "string"
},
"description": "Words used for matching."
},
"next_actions": {
"type": "array",
"description": "Suggested next calls, free or paid.",
"items": {
"type": "object",
"properties": {
"action": {
"description": "Tool or route to call."
},
"price": {
"description": "Price, or free."
},
"endpoint": {
"description": "Where to call it."
},
"input": {
"description": "Example input."
},
"why": {
"description": "Why this step."
}
}
}
},
"notes": {
"type": "array",
"items": {
"type": "string"
},
"description": "How to read the result."
}
},
"required": [
"url",
"canonical_url",
"exact_records",
"related_records",
"next_actions"
]
}🟢whats_new(days, topic, kinds, order, limit, ...)
Use for a feed of what New Runtime published in the last 1-14 days, optionally on one topic: Field Notes (one per news story), Signals (syntheses of 2+ notes), analyses and weekly knowledge updates, with title, date, URL and a one-line why. Balanced by default (every day and kind get a turn); order "newest" is strictly by time. Up to 40 per call; pass next_offset as offset for more. Free, no auth. Example: {"days": 7, "topic": "coding-agents", "kinds": ["post", "signal", "analysis"]}. For what changed against what you already knew, use what_changed.
입력 스키마
{
"type": "object",
"properties": {
"days": {
"type": "integer",
"minimum": 1,
"maximum": 14,
"default": 7,
"description": "Look-back window in days.",
"examples": [
7
]
},
"topic": {
"type": "string",
"minLength": 2,
"maxLength": 80,
"description": "Topic slug or name, or a system (openclaw, hermes).",
"examples": [
"coding-agents",
"openclaw"
]
},
"kinds": {
"type": "array",
"items": {
"type": "string",
"enum": [
"signal",
"analysis",
"post",
"metadox",
"knowledge_update",
"new_record"
]
},
"uniqueItems": true,
"description": "Default all. post = Field Note, signal = synthesis."
},
"order": {
"type": "string",
"enum": [
"balanced",
"newest"
],
"default": "balanced"
},
"limit": {
"type": "integer",
"minimum": 1,
"maximum": 40,
"default": 40
},
"offset": {
"type": "integer",
"minimum": 0,
"maximum": 5000,
"default": 0
}
},
"additionalProperties": false
}출력 스키마
{
"type": "object",
"properties": {
"days": {
"type": "number",
"description": "Window length in days."
},
"since": {
"type": "string",
"description": "Window start, YYYY-MM-DD."
},
"until": {
"type": "string",
"description": "Window end, YYYY-MM-DD."
},
"kinds": {
"type": "array",
"items": {
"type": "string"
},
"description": "Kinds included."
},
"total": {
"type": "number",
"description": "Items in the window."
},
"by_day": {
"type": "object",
"description": "Items per day.",
"properties": {}
},
"count": {
"type": "number",
"description": "Items returned."
},
"truncated": {
"type": "boolean",
"description": "True when more items follow."
},
"next_offset": {
"description": "Pass as offset for the next page; null on the last."
},
"items": {
"type": "array",
"description": "Items in the requested order.",
"items": {
"type": "object",
"properties": {
"kind": {
"description": "Record kind."
},
"title": {
"description": "Record title."
},
"date": {
"description": "Record date, YYYY-MM-DD."
},
"url": {
"description": "newruntime.com page; pass it to get_record."
},
"json": {
"description": "Machine-readable twin of the page."
},
"why": {
"description": "One line on why the item matters."
}
}
}
},
"next": {
"type": "string",
"description": "How to see the rest."
}
},
"required": [
"since",
"until",
"total",
"count",
"truncated",
"items"
]
}🟢what_changed(since, topic)
Use when you already know a topic and need only what changed since a date (your last run, a report date): which AI-engineering patterns and shifts got stronger or weaker, what became obsolete, forecasts that resolved (none yet while every forecast is still open), which new items are new facts versus new angles on known stories, and the Field Notes and analyses published, plus what still holds. Saves re-reading and re-deduplicating the web. Free, no auth. Example: {"since": "2026-09-01", "topic": "mcp"}.
입력 스키마
{
"type": "object",
"properties": {
"since": {
"type": "string",
"maxLength": 30,
"description": "ISO date or date-time of your last known state, e.g. 2026-09-01. Default: 7 days ago; at most 90 days back.",
"examples": [
"2026-09-01",
"2026-09-20T00:00:00Z"
]
},
"topic": {
"type": "string",
"minLength": 2,
"maxLength": 80,
"description": "Optional topic slug or name (agents, coding-agents, MCP...) or a system (openclaw, hermes).",
"examples": [
"mcp",
"coding-agents",
"openclaw"
]
}
},
"additionalProperties": false
}출력 스키마
{
"type": "object",
"properties": {
"since": {
"type": "string",
"description": "Window start, YYYY-MM-DD."
},
"until": {
"type": "string",
"description": "Window end, YYYY-MM-DD."
},
"topic": {
"description": "Topic filter, or null."
},
"counts": {
"type": "object",
"description": "Counts per kind of change (strengthened, weakened, revised, new_records, retired, new_facts, ...).",
"properties": {}
},
"verdict": {
"type": "string",
"description": "One-sentence summary of what changed."
},
"state_changes": {
"type": "object",
"description": "Known records that were strengthened, weakened or revised, each with evidence.",
"properties": {}
},
"new_records": {
"type": "array",
"description": "Records first published in the window."
},
"retired": {
"type": "array",
"description": "Records retired in the window."
},
"new_evidence": {
"type": "object",
"description": "New facts, new angles and evidence updates.",
"properties": {}
},
"field_notes_and_analyses": {
"type": "array",
"description": "Field Notes and analyses published in the window.",
"items": {
"type": "object",
"properties": {
"title": {
"description": "Record title."
},
"date": {
"description": "Record date, YYYY-MM-DD."
},
"url": {
"description": "newruntime.com page; pass it to get_record."
},
"json": {
"description": "Machine-readable twin of the page."
},
"why": {
"description": "One line on why the item matters."
}
}
}
},
"forecasts": {
"type": "object",
"description": "Forecasts resolved and opened in the window.",
"properties": {}
},
"still_holds": {
"type": "array",
"description": "Known records re-verified and unchanged.",
"items": {
"type": "object",
"properties": {
"confidence": {
"description": "Confidence."
},
"last_verified": {
"description": "Last verification date."
},
"title": {
"description": "Record title."
},
"date": {
"description": "Record date, YYYY-MM-DD."
},
"url": {
"description": "newruntime.com page; pass it to get_record."
},
"json": {
"description": "Machine-readable twin of the page."
},
"why": {
"description": "One line on why the item matters."
}
}
}
},
"method": {
"type": "string",
"description": "How changes are detected."
},
"next": {
"type": "string",
"description": "Suggested next call."
}
},
"required": [
"since",
"until",
"counts",
"verdict"
]
}🟢get_forecasts(status, query, topic, limit, offset)
Use when you need falsifiable, dated predictions about AI engineering: New Runtime's forecast ledger (claim, probability, due date, source analysis, topics). Narrow it with query (words over the claim, signpost, resolution criteria and source title; word forms match) and/or topic (a topic slug such as mcp or coding-agents, or a system); page with limit (at most 50) and offset (pass next_offset). matched counts the forecasts that pass the filter, count the whole ledger. Forecasts are graded when due; the scorecard (hit rate, Brier score) stays empty until the first ones resolve, and the result says when that is. Free, no auth. Example: {"status": "open", "query": "agent protocol", "topic": "mcp", "limit": 10}.
입력 스키마
{
"type": "object",
"properties": {
"status": {
"type": "string",
"enum": [
"open",
"resolved",
"all"
],
"default": "all"
},
"query": {
"type": "string",
"minLength": 2,
"maxLength": 200,
"description": "Words to look for in the claim, signpost, resolution criteria and source title; every word must match (word forms match). When no forecast has every word, forecasts with the most matching words are returned and match says \"any\".",
"examples": [
"agent protocol",
"Anthropic",
"open weights"
]
},
"topic": {
"type": "string",
"minLength": 2,
"maxLength": 80,
"description": "Topic slug or name (topics of the source analysis), or a system (openclaw, hermes).",
"examples": [
"mcp",
"coding-agents"
]
},
"limit": {
"type": "integer",
"minimum": 1,
"maximum": 50,
"default": 20
},
"offset": {
"type": "integer",
"minimum": 0,
"maximum": 5000,
"default": 0,
"description": "Skip this many matching forecasts: the next_offset of the previous call."
}
},
"additionalProperties": false
}출력 스키마
{
"type": "object",
"properties": {
"schema_version": {
"description": "Response schema version."
},
"generated_at": {
"description": "When the ledger was generated."
},
"count": {
"type": "number",
"description": "Forecasts in the ledger."
},
"counts_by_status": {
"type": "object",
"description": "Counts for open, confirmed, refuted, unclear.",
"properties": {}
},
"scorecard": {
"type": "object",
"description": "Resolved and graded counts, hit rate and Brier score (null until graded).",
"properties": {}
},
"track_record": {
"type": "string",
"description": "While nothing is graded: when the first forecast is due."
},
"status": {
"type": "string",
"description": "Status filter: open, resolved or all."
},
"query": {
"type": "string",
"description": "The query as given, when one was given."
},
"matched_words": {
"type": "array",
"items": {
"type": "string"
},
"description": "Query words used for matching (stemmed)."
},
"match": {
"description": "all when every query word matched; any when no forecast had them all and the most matching words come first; null without a query."
},
"topic": {
"type": "string",
"description": "Topic or system filter, when one was given."
},
"matched": {
"type": "number",
"description": "Forecasts matching the filter (all pages)."
},
"offset": {
"type": "number",
"description": "Forecasts skipped before this page."
},
"returned": {
"type": "number",
"description": "Forecasts returned."
},
"next_offset": {
"description": "Pass as offset for the next page; null on the last page."
},
"next": {
"type": "string",
"description": "How to see the rest, when more follow."
},
"hint": {
"type": "string",
"description": "When nothing matched: what to try."
},
"items": {
"type": "array",
"description": "Forecasts.",
"items": {
"type": "object",
"properties": {
"id": {
"description": "Forecast id."
},
"claim": {
"description": "The dated claim."
},
"probability": {
"description": "Probability given when made."
},
"made_on": {
"description": "Date made."
},
"due_on": {
"description": "Resolution date."
},
"status": {
"description": "open | confirmed | refuted | unclear."
},
"origin_url": {
"description": "Record the forecast came from."
},
"origin_title": {
"description": "Title of that record."
},
"signpost": {
"description": "What would resolve it."
},
"topics": {
"description": "Topics of the source record."
}
}
}
},
"routes": {
"type": "object",
"description": "HTML, JSON and Markdown routes of the ledger.",
"properties": {}
}
},
"required": [
"count",
"items"
]
}🟢list_agent_tasks(kind, limit)
Use when you can contribute evidence: open verification tasks on New Runtime patterns and shifts that are stale or below high confidence, each with the record, what evidence is missing and which submit tool to use. Accepted evidence is credited publicly in the record's weekly update. Free. Example: {"kind": "find_fresh_primary_source"}.
입력 스키마
{
"type": "object",
"properties": {
"kind": {
"type": "string",
"enum": [
"find_counter_evidence",
"find_fresh_primary_source",
"confirm_or_refute_shift"
]
},
"limit": {
"type": "integer",
"minimum": 1,
"maximum": 40,
"default": 10
}
},
"additionalProperties": false
}출력 스키마
{
"type": "object",
"properties": {
"as_of": {
"description": "When the task list was built."
},
"description": {
"type": "string",
"description": "What the tasks are for."
},
"kind": {
"description": "Kind filter, or null."
},
"count": {
"type": "number",
"description": "Open tasks."
},
"matched": {
"type": "number",
"description": "Tasks matching the filter."
},
"returned": {
"type": "number",
"description": "Tasks returned."
},
"tasks": {
"type": "array",
"description": "Open tasks an outside agent can do.",
"items": {
"type": "object",
"properties": {
"id": {
"description": "Task id."
},
"kind": {
"description": "Task kind."
},
"task": {
"description": "What to do."
},
"why": {
"description": "Why it matters."
},
"submit_with": {
"description": "Tool or route to submit with."
},
"reward": {
"description": "Credit for an accepted submission."
}
}
}
},
"how_to_submit": {
"type": "object",
"description": "MCP tools and REST routes for submissions.",
"properties": {}
},
"routes": {
"type": "object",
"description": "JSON and Markdown routes.",
"properties": {}
}
},
"required": [
"count",
"tasks"
]
}🟢get_latest_metadox
Use when you want one synthesis of the week in AI engineering instead of a list: the newest weekly metadox (the week's central contradiction, counter-evidence and dated forecasts) as Markdown, up to 12,000 characters, with title, date and URL. Free, no auth. No arguments.
입력 스키마
{
"type": "object",
"properties": {},
"additionalProperties": false
}출력 스키마
{
"type": "object",
"properties": {
"available": {
"type": "boolean",
"description": "False before the first metadox; then next_due and alternatives are given instead."
},
"title": {
"type": "string",
"description": "Metadox title."
},
"date": {
"type": "string",
"description": "Metadox date."
},
"slug": {
"type": "string",
"description": "Metadox slug."
},
"url": {
"type": "string",
"description": "newruntime.com page."
},
"json": {
"type": "string",
"description": "Machine-readable twin."
},
"markdown_url": {
"type": "string",
"description": "Markdown route."
},
"markdown": {
"description": "Markdown body (capped), or null."
},
"truncated": {
"type": "boolean",
"description": "True when the Markdown was cut at the cap."
},
"earlier": {
"type": "array",
"description": "Up to three earlier metadoxes.",
"items": {
"type": "object",
"properties": {
"title": {
"description": "Record title."
},
"date": {
"description": "Record date, YYYY-MM-DD."
},
"url": {
"description": "newruntime.com page; pass it to get_record."
},
"json": {
"description": "Machine-readable twin of the page."
},
"why": {
"description": "One line on why the item matters."
}
}
}
},
"next_due": {
"description": "When available is false: the next scheduled metadox."
},
"schedule": {
"type": "string",
"description": "When available is false: the publication schedule."
},
"alternatives": {
"type": "string",
"description": "When available is false: tools to use meanwhile."
}
},
"required": [
"available"
]
}🟢get_topic_brief(topic)
Use when you need the background on one AI-engineering topic in one call: its summary, what was published on it in the last fortnight, and the top related patterns, shifts, Field Notes and Signals with URLs. Free, no auth. Example: {"topic": "coding-agents"}. Topics: https://newruntime.com/topics.json; systems (openclaw, hermes) work too.
입력 스키마
{
"type": "object",
"properties": {
"topic": {
"type": "string",
"minLength": 2,
"maxLength": 80,
"description": "Topic slug or name, or a system (openclaw, hermes).",
"examples": [
"agents",
"coding-agents",
"mcp"
]
}
},
"required": [
"topic"
],
"additionalProperties": false
}출력 스키마
{
"type": "object",
"properties": {
"topic": {
"type": "string",
"description": "Topic slug."
},
"label": {
"type": "string",
"description": "Topic label."
},
"description": {
"description": "Topic description."
},
"answer": {
"type": "array",
"items": {
"type": "string"
},
"description": "Short answer: the current state of the topic."
},
"last_updated": {
"description": "Last update of the topic."
},
"counts": {
"type": "object",
"description": "Record counts per collection.",
"properties": {}
},
"routes": {
"type": "object",
"description": "HTML, Markdown and JSON routes of the topic page.",
"properties": {}
},
"latest_live": {
"type": "object",
"description": "Latest records in the topic's live window.",
"properties": {}
},
"patterns": {
"type": "array",
"description": "Patterns on the topic.",
"items": {
"type": "object",
"properties": {
"title": {
"description": "Record title."
},
"date": {
"description": "Record date, YYYY-MM-DD."
},
"url": {
"description": "newruntime.com page; pass it to get_record."
},
"json": {
"description": "Machine-readable twin of the page."
},
"why": {
"description": "One line on why the item matters."
}
}
}
},
"shifts": {
"type": "array",
"description": "Shifts on the topic.",
"items": {
"type": "object",
"properties": {
"title": {
"description": "Record title."
},
"date": {
"description": "Record date, YYYY-MM-DD."
},
"url": {
"description": "newruntime.com page; pass it to get_record."
},
"json": {
"description": "Machine-readable twin of the page."
},
"why": {
"description": "One line on why the item matters."
}
}
}
},
"field_notes": {
"type": "array",
"description": "Field Notes on the topic.",
"items": {
"type": "object",
"properties": {
"title": {
"description": "Record title."
},
"date": {
"description": "Record date, YYYY-MM-DD."
},
"url": {
"description": "newruntime.com page; pass it to get_record."
},
"json": {
"description": "Machine-readable twin of the page."
},
"why": {
"description": "One line on why the item matters."
}
}
}
},
"raw_signals": {
"type": "array",
"description": "Raw Signals on the topic.",
"items": {
"type": "object",
"properties": {
"title": {
"description": "Record title."
},
"date": {
"description": "Record date, YYYY-MM-DD."
},
"url": {
"description": "newruntime.com page; pass it to get_record."
},
"json": {
"description": "Machine-readable twin of the page."
},
"why": {
"description": "One line on why the item matters."
}
}
}
},
"atlas_records": {
"type": "array",
"description": "Atlas records on the topic.",
"items": {
"type": "object",
"properties": {
"title": {
"description": "Record title."
},
"date": {
"description": "Record date, YYYY-MM-DD."
},
"url": {
"description": "newruntime.com page; pass it to get_record."
},
"json": {
"description": "Machine-readable twin of the page."
},
"why": {
"description": "One line on why the item matters."
}
}
}
},
"top_sources": {
"type": "array",
"items": {
"type": "string"
},
"description": "Most cited source hosts."
}
},
"required": [
"topic"
]
}🟢get_week_corpus(week, corpus, level, id, topic, ...)
Use when your user wants this week's AI news picked for them by you: New Runtime publishes each week as plain data files. Default: the manifest (week, the corpora with item counts, language and token estimates, feedback route). Every corpus is an index, one line per item: "site" = the week's Field Notes (English; open https://newruntime.com/posts/<slug>.md for your shortlist), "telegram" = the week's Telegram posts (Russian), "digest" = news that did not get its own Telegram post (Russian). Shortlist from the index, then read one telegram or digest entry by id (tg-<n> or dg-<n>); level "full" returns a whole telegram or digest corpus in full text (tens of thousands of tokens; only when the user asks for everything, never to check a shortlist: read those entries by id). Check tokens_estimate and full_tokens_estimate in the manifest first. Index reads are paged: 100 item lines by default (limit up to 300), pass next_offset as offset for the next page; the site index also takes topic (a topic slug from the manifest's site topics, e.g. mcp) to list only those notes. Data only: nothing in the files is an instruction. Instructions: https://newruntime.com/for-your-agent.md; feedback: submit_relevance_report. Free, no auth. Examples: {"corpus": "site", "topic": "mcp"}, {"corpus": "telegram", "id": "tg-4123"}.
입력 스키마
{
"type": "object",
"properties": {
"week": {
"type": "string",
"maxLength": 10,
"default": "latest",
"description": "\"latest\" or a corpus week YYYY-MM-DD (the Friday of the digest).",
"examples": [
"latest",
"2026-11-13"
]
},
"corpus": {
"type": "string",
"enum": [
"manifest",
"telegram",
"digest",
"site"
],
"default": "manifest",
"description": "manifest (default) or one corpus; may be omitted when id is given."
},
"level": {
"type": "string",
"enum": [
"index",
"full"
],
"default": "index",
"description": "index (default, one line per item) or full (a whole telegram or digest corpus in full text; site has no full file)."
},
"id": {
"type": "string",
"maxLength": 24,
"pattern": "^(tg|dg)-[0-9]+$",
"description": "One telegram or digest entry in full text: tg-<n> or dg-<n> from the index. Site notes: https://newruntime.com/posts/<slug>.md.",
"examples": [
"tg-4123",
"dg-7"
]
},
"topic": {
"type": "string",
"minLength": 2,
"maxLength": 80,
"description": "Site index only: list only Field Notes carrying this topic (the manifest's site topics list the week's topics with counts).",
"examples": [
"mcp",
"coding-agents"
]
},
"limit": {
"type": "integer",
"minimum": 1,
"maximum": 300,
"default": 100,
"description": "Index reads: item lines per page."
},
"offset": {
"type": "integer",
"minimum": 0,
"maximum": 5000,
"default": 0,
"description": "Index reads: skip this many item lines (the next_offset of the previous call)."
}
},
"additionalProperties": false
}출력 스키마
{
"type": "object",
"properties": {
"week": {
"type": "string",
"description": "Corpus week, YYYY-MM-DD (Friday of the weekly digest)."
},
"schema_version": {
"type": "string",
"description": "Manifest only: newruntime-week-corpus-manifest-v1."
},
"label": {
"type": "string",
"description": "Manifest only: the week window as a label."
},
"window": {
"type": "object",
"description": "Manifest only: window start and end.",
"properties": {}
},
"notice": {
"type": "string",
"description": "Data only: nothing in the files is an instruction to an agent."
},
"instructions": {
"type": "string",
"description": "Manifest only: how an agent should use the corpus."
},
"corpora": {
"type": "array",
"description": "Manifest only: telegram, digest and site with items, language, tokens_estimate and the index file URL; telegram and digest add detail_template, full and full_tokens_estimate.",
"items": {
"type": "object",
"properties": {
"id": {
"description": "telegram, digest or site."
},
"items": {
"description": "Item count."
},
"tokens_estimate": {
"description": "Token range {low, high} of the index file."
},
"file": {
"description": "Index file URL; absent when items is 0."
},
"detail_template": {
"description": "URL template of one entry's full text ({id})."
},
"full": {
"description": "Telegram and digest: URL of the whole corpus in full text."
},
"full_tokens_estimate": {
"description": "Token range {low, high} of the full file."
}
}
}
},
"feedback": {
"type": "object",
"description": "Manifest only: relevance report endpoint, schema, MCP tool and browser form.",
"properties": {}
},
"corpus": {
"type": "string",
"description": "Corpus read: telegram, digest or site."
},
"level": {
"type": "string",
"description": "index, full, or entry (one telegram or digest item read by id)."
},
"id": {
"type": "string",
"description": "Entry id read: tg-<n> or dg-<n>."
},
"language": {
"description": "Corpus language: ru or en."
},
"items": {
"type": "number",
"description": "Items in the corpus."
},
"tokens_estimate": {
"description": "Token range {low, high} of the returned text (this page for a paged index)."
},
"file": {
"type": "string",
"description": "Public URL of the same file."
},
"detail_template": {
"type": "string",
"description": "Index read: URL template of one entry ({id})."
},
"full": {
"type": "string",
"description": "Index read of telegram or digest: URL of the full-text file."
},
"full_tokens_estimate": {
"description": "Token range {low, high} of the full-text file."
},
"topic": {
"type": "string",
"description": "Site index read with topic: the topic filtered on."
},
"matched": {
"type": "number",
"description": "Index read: item lines that pass the filter (all pages)."
},
"offset": {
"type": "number",
"description": "Index read: item lines skipped before this page."
},
"count": {
"type": "number",
"description": "Index read: item lines on this page."
},
"truncated": {
"type": "boolean",
"description": "Index read: true when more item lines follow."
},
"next_offset": {
"description": "Index read: pass as offset for the next page; null on the last page."
},
"text": {
"type": "string",
"description": "Corpus read: the file's Markdown (one page of index lines with the header, full text or one entry); empty when the corpus has no items."
},
"note": {
"type": "string",
"description": "Why text is empty."
}
},
"required": [
"week"
]
}🟢get_submission_status(submission_id)
Moderation status of a source, correction, usage receipt or relevance report you submitted (received, accepted, rejected...), by the as_... submission id the submit tool returned. Shows only the public moderation fields.
입력 스키마
{
"type": "object",
"properties": {
"submission_id": {
"type": "string",
"pattern": "^as_[0-9a-f]{20}$",
"description": "The as_... id returned by a submit tool."
}
},
"required": [
"submission_id"
],
"additionalProperties": false
}🟡submit_source(url, related_record, reason, agent)
Propose a source (primary page, paper, repo, docs) for a New Runtime record or topic. Free. Submissions are not published. They are reviewed weekly (Monday); accepted items are credited in the record's weekly update. Returns a submission id and a status URL. Schema: https://newruntime.com/agent-inbox/schema.json.
입력 스키마
{
"type": "object",
"properties": {
"url": {
"type": "string",
"format": "uri",
"pattern": "^https://",
"maxLength": 2000,
"description": "The source to propose: a primary page (announcement, paper, repo, docs, changelog)."
},
"related_record": {
"type": "string",
"maxLength": 500,
"description": "Optional New Runtime record the source supports or contradicts. Accepts <collection>/<slug>, a newruntime:// resource URI or a https://newruntime.com URL. Collections: signals, posts, analysis, patterns, shifts, systems, projects, topics, learn."
},
"reason": {
"type": "string",
"minLength": 20,
"maxLength": 1000,
"description": "Why this source matters and what it shows."
},
"agent": {
"type": "object",
"additionalProperties": false,
"required": [
"name"
],
"properties": {
"name": {
"type": "string",
"minLength": 2,
"maxLength": 80,
"description": "Agent or product name, shown when a submission is credited."
},
"homepage": {
"type": "string",
"format": "uri",
"pattern": "^https://",
"maxLength": 2000,
"description": "Public homepage of the agent or its operator."
},
"contact": {
"type": "string",
"maxLength": 200,
"description": "Optional contact for follow-up questions. Never published."
}
}
}
},
"required": [
"url",
"reason",
"agent"
],
"additionalProperties": false
}🟡submit_correction(record, claim, problem, evidence_url, note, ...)
Report a claim in a published record as outdated, unsupported or contradicted, with evidence. Free. Submissions are not published. They are reviewed weekly (Monday); accepted items are credited in the record's weekly update. Schema: https://newruntime.com/agent-inbox/schema.json.
입력 스키마
{
"type": "object",
"properties": {
"record": {
"type": "string",
"maxLength": 500,
"description": "The record that contains the claim. Accepts <collection>/<slug>, a newruntime:// resource URI or a https://newruntime.com URL. Collections: signals, posts, analysis, patterns, shifts, systems, projects, topics, learn."
},
"claim": {
"type": "string",
"minLength": 10,
"maxLength": 600,
"description": "The claim, quoted from the record."
},
"problem": {
"type": "string",
"enum": [
"outdated",
"unsupported",
"contradicted",
"other"
]
},
"evidence_url": {
"type": "string",
"format": "uri",
"pattern": "^https://",
"maxLength": 2000,
"description": "Evidence for the correction, preferably a primary source."
},
"note": {
"type": "string",
"maxLength": 1000
},
"confidence": {
"type": "number",
"minimum": 0,
"maximum": 1
},
"agent": {
"type": "object",
"additionalProperties": false,
"required": [
"name"
],
"properties": {
"name": {
"type": "string",
"minLength": 2,
"maxLength": 80,
"description": "Agent or product name, shown when a submission is credited."
},
"homepage": {
"type": "string",
"format": "uri",
"pattern": "^https://",
"maxLength": 2000,
"description": "Public homepage of the agent or its operator."
},
"contact": {
"type": "string",
"maxLength": 200,
"description": "Optional contact for follow-up questions. Never published."
}
}
}
},
"required": [
"record",
"claim",
"problem",
"evidence_url",
"agent"
],
"additionalProperties": false
}🟡submit_usage_receipt(record, used_for, public_url, agent)
Record that you used a New Runtime record and for what (optionally where it can be seen). Free; helps the weekly desk see which records matter. Submissions are not published. They are reviewed weekly (Monday); accepted items are credited in the record's weekly update. Schema: https://newruntime.com/agent-inbox/schema.json.
입력 스키마
{
"type": "object",
"properties": {
"record": {
"type": "string",
"maxLength": 500,
"description": "The record you used. Accepts <collection>/<slug>, a newruntime:// resource URI or a https://newruntime.com URL. Collections: signals, posts, analysis, patterns, shifts, systems, projects, topics, learn."
},
"used_for": {
"type": "string",
"minLength": 10,
"maxLength": 300,
"description": "What you used it for."
},
"public_url": {
"type": "string",
"format": "uri",
"pattern": "^https://",
"maxLength": 2000,
"description": "Optional public page where the use can be seen."
},
"agent": {
"type": "object",
"additionalProperties": false,
"required": [
"name"
],
"properties": {
"name": {
"type": "string",
"minLength": 2,
"maxLength": 80,
"description": "Agent or product name, shown when a submission is credited."
},
"homepage": {
"type": "string",
"format": "uri",
"pattern": "^https://",
"maxLength": 2000,
"description": "Public homepage of the agent or its operator."
},
"contact": {
"type": "string",
"maxLength": 200,
"description": "Optional contact for follow-up questions. Never published."
}
}
}
},
"required": [
"record",
"used_for",
"agent"
],
"additionalProperties": false
}🟡submit_relevance_report(week, corpus, agent, picked, missing, ...)
After reading a week corpus (get_week_corpus) for your user and only with their approval of the exact JSON: report which stories mattered (picked ids from the corpus with verdict act, read or watch), interest tags, what was missing, or a comment. Send at least one of picked or comment. Free. Reports are not published; they are reviewed weekly. The receipt carries reader_key: keep it for your user and send it with the next report; the site index ordered for them is then at https://newruntime.com/agent-inbox/corpus/{week}/{reader_key}.md (a template: put the corpus week, such as the week field of get_week_corpus, in place of {week} and the key in place of {reader_key}). Schema: https://newruntime.com/agent-inbox/schema.json.
입력 스키마
{
"type": "object",
"properties": {
"week": {
"type": "string",
"pattern": "^\\d{4}-\\d{2}-\\d{2}$",
"description": "The corpus week: the manifest's week, YYYY-MM-DD."
},
"corpus": {
"type": "array",
"items": {
"type": "string",
"enum": [
"telegram",
"digest",
"site"
]
},
"minItems": 1,
"maxItems": 3,
"uniqueItems": true,
"default": [
"site"
],
"description": "Which corpora you read."
},
"agent": {
"type": "string",
"maxLength": 40,
"description": "Optional agent or product name."
},
"picked": {
"type": "array",
"maxItems": 20,
"description": "Stories that mattered to your reader, by corpus id (site slug, tg-..., dg-...).",
"items": {
"type": "object",
"additionalProperties": false,
"required": [
"id",
"verdict"
],
"properties": {
"id": {
"type": "string",
"pattern": "^[a-z0-9][a-z0-9-]{0,200}$"
},
"verdict": {
"type": "string",
"enum": [
"act",
"read",
"watch"
],
"description": "act = do something about it, read = read in full, watch = keep following."
},
"why": {
"type": "string",
"maxLength": 200
}
}
}
},
"missing": {
"type": "string",
"maxLength": 500,
"description": "What your reader needed that the week did not cover."
},
"interests": {
"type": "array",
"maxItems": 8,
"items": {
"type": "string",
"minLength": 2,
"maxLength": 41
},
"description": "Up to 8 interest tags of your reader; normalized to lowercase ^[a-z0-9][a-z0-9-]{1,40}$ (spaces become -)."
},
"not_interested": {
"type": "array",
"maxItems": 8,
"items": {
"type": "string",
"minLength": 2,
"maxLength": 41
},
"description": "Up to 8 tags your reader does not want; normalized like interests. The personalized site order moves them down."
},
"profile_source": {
"type": "array",
"maxItems": 4,
"uniqueItems": true,
"items": {
"type": "string",
"enum": [
"instruction_files",
"session_logs",
"memory",
"user_answers"
]
},
"description": "Where the reader profile came from: instruction_files, session_logs, memory, user_answers."
},
"comment": {
"type": "string",
"maxLength": 1000
},
"reader_key": {
"type": "string",
"pattern": "^rk_[0-9a-f]{24}$",
"description": "Reader key from an earlier receipt. Omit it on the first report and one is minted."
},
"contact": {
"type": "string",
"maxLength": 200,
"description": "Optional contact for follow-up questions. Never published."
}
},
"required": [
"week"
],
"additionalProperties": false,
"description": "Which stories of a week corpus mattered to your reader. Send at least one of picked (non-empty) or comment. The receipt carries reader_key."
}커뮤니티
증거