Search / com.newruntime/archive
New Runtime public records
R0· no sign-inAI-engineering archive: search, what is new, forecasts, topic briefs, URL context; moderated inbox.
v0.7.0 · active · website · descriptor JSON · versions
Search / com.newruntime/archive
R0· no sign-inAI-engineering archive: search, what is new, forecasts, topic briefs, URL context; moderated inbox.
v0.7.0 · active · website · descriptor JSON · versions
{
"mcpServers": {
"archive": {
"type": "http",
"url": "https://newruntime.com/mcp"
}
}
}claude mcp add --transport http archive https://newruntime.com/mcp[mcp_servers.archive]
url = "https://newruntime.com/mcp"
{
"$schema": "https://opencode.ai/config.json",
"mcp": {
"archive": {
"type": "remote",
"url": "https://newruntime.com/mcp",
"enabled": true
}
}
}{
"mcpServers": {
"archive": {
"type": "http",
"url": "https://newruntime.com/mcp"
}
},
"deeplink": "cursor://anysphere.cursor-deeplink/mcp/install?name=archive&config=eyJ0eXBlIjoiaHR0cCIsInVybCI6Imh0dHBzOi8vbmV3cnVudGltZS5jb20vbWNwIn0%3D"
}{
"servers": {
"archive": {
"type": "http",
"url": "https://newruntime.com/mcp"
}
}
}{
"mcpServers": {
"archive": {
"httpUrl": "https://newruntime.com/mcp"
}
}
}extensions:
"archive":
enabled: true
name: "archive"
type: streamable_http
uri: "https://newruntime.com/mcp"
timeout: 300
protocol.modernsupports 2026-07-28 (server/discover)protocol.statelessanswers without a sessionprotocol.transportstreamable HTTPprotocol.list_ttltools/list carries no ttlMs, so clients re-fetch itauth.prm, auth.as_metadata, auth.cimdno remote uses OAuthauth.secret_in_urlno templated URLtools.descriptionsevery tool has a descriptiontools.description_lengthdescriptions are concisetools.schemasevery tool has an object input schematools.annotations14 of 14 tools declare readOnlyHint or destructiveHinttools.directory_hintsevery tool declares readOnlyHint, destructiveHint, idempotentHint and openWorldHinttools.token_costabout 3,164 tokens to load every tooltools.api_dumptools are not a one-to-one API dumpstability.changes17 tool changes in 30 daysstability.rug_pull8 tools kept their name but changed most of their description in 30 days; review before trusting themdeps.known_vulns, deps.mcp_sdk_version, deps.resolvableno npm or PyPI packageget_topic_briefchangeddescription, input schemameaning changedA topic hub in one call: description and answer, counts, what was published on it in the last fortnight, and the top related patterns, shifts, field notes and signals. Accepts a topic slug or name (for example agents, coding-agents, MCP).
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 news signals with URLs. Free, no auth. Example: {"topic": "coding-agents"}. Topic list: https://newruntime.com/topics.json.
get_latest_metadoxchangeddescriptionmeaning changedThe newest weekly metadox (New Runtime's Monday synthesis of the week's central contradiction, with forecasts): title, date, URL and its Markdown text (truncated at 12,000 characters). Says when the next one is due if none is published yet.
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.
list_agent_taskschangeddescriptionVerification work an outside agent can pick up: New Runtime patterns and shifts that are stale or below high confidence, each with the record, why it needs evidence and where to submit (submit_source or submit_correction). Accepted evidence is credited in the record's weekly update.
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"}.
get_forecastschangeddescriptionmeaning changedDated, probability-weighted forecasts made by New Runtime analyses and weekly metadoxes, with the public scorecard (hit rate, Brier score). Open forecasts come soonest-due first; resolved ones carry the grade note and evidence.
Use when you need falsifiable predictions about AI engineering with a track record: New Runtime's dated forecasts (claim, probability, due date, source analysis) and the public scorecard (hit rate, Brier score); resolved ones carry the grade and evidence. Free, no auth. Example: {"status": "open", "limit": 10}.
what_changedchangeddescription, input schemameaning changedState changes of the New Runtime knowledge base since a date (default 7 days, up to 90), optionally for one topic: patterns and shifts the weekly desk strengthened or weakened, new records, practices moved to the graveyard (obsolete), forecasts resolved, and new evidence split into new facts versus new angles on known stories. Also lists the known context that did not change. Use it instead of re-reading the web to learn what moved since your last visit.
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, which forecasts resolved, and which new items are new facts versus new angles on known stories, plus what still holds. Saves re-reading and re-deduplicating the web. Free, no auth. Example: {"since": "2026-09-01", "topic": "mcp"}.
whats_newchangeddescription, input schemameaning changedNewest-first list of what New Runtime published in the last 1-14 days: raw signals, analyses, field notes, the weekly metadox, weekly knowledge updates to patterns/shifts/systems/projects/deltas/graveyard/learn guides, and records the weekly desk created. Each item has kind, title, date, url, json and why. At most 40 items.
Use for a quick feed of what New Runtime published in the last 1-14 days (news signals, analyses, field notes, weekly knowledge updates), newest first, at most 40 items with title, date, URL and a one-line why. Free, no auth. Example: {"days": 7, "kinds": ["signal", "analysis"]}. For what changed against what you already knew, use what_changed.
contextualize_urlchangeddescription, input schemameaning changedFree first step when researching any AI/software URL: New Runtime records that cite it (exact source match), related records and patterns (keyword match on the URL and your task), the source-origin hub, and next actions (paid prior-coverage check, research bundle, primary source; free source submission). Returns no coverage verdict.
Use first when you found an AI or software URL elsewhere (a release note, repo, blog post, tweet) and want to know if it is new: returns New Runtime records that cite that exact URL, related records and patterns, the source's hub, and next steps. Free, no auth. Example: {"url": "https://github.com/modelcontextprotocol/servers", "task": "is this new?"}.
list_paid_toolschangeddescriptionDescribe the pay-per-call x402 tools (primary-source resolver, rumour check, research bundle, prior-coverage check): endpoint, price in USDC, input example, when a call is charged, and how an agent pays without an account.
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.
get_recordchangeddescription, input schemameaning changedRead the full structured record (title, dates, sources, body) of a Field Note, Raw Signal or Analysis by kind and slug, or by its newruntime.com URL.
Use after search_archive, whats_new or what_changed to read one record in full: title, dates, source URLs, topics and body text. Pass the newruntime.com URL you got, or kind + slug. Free, no auth. Example: {"url": "https://newruntime.com/signals/<slug>/"}.
search_archivechangeddescription, input schemameaning changedKeyword search over published Field Notes (analysis), Raw Signals (source-linked news) and authorial Analysis about AI engineering, agents and developer tools. Returns titles, dates, descriptions and canonical URLs, newest-published first among equal scores.
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. Searches New Runtime's Field Notes, Raw Signals and Analyses; returns up to 10 hits with title, date, one-line summary and URL. Free, no auth. Example: {"query": "MCP security", "limit": 5}. Then get_record for the full text.
Showing the latest 10. Full history: list_changes · trend: quality JSON.
Checks run on what our credential-free, read-only probes observe; tools are never called and no code audit is performed. How it is computed.
+dns-namespace +website +reachable +uptime-100% +updated-22d-ago -no-repository
Derived from observable signals (official registry feed and our own credential-free probes); no code audit performed. How it is computed.
| remote | auth | reachable | uptime 30d | p50 | protocol | last ok |
|---|---|---|---|---|---|---|
https://newruntime.com/mcp streamable-http | none | yes | 100% | 408 ms | 2026-07-28 | 2026-10-19 |
contextualize_urlread-only | Use first when you found an AI or software URL elsewhere (a release note, repo, blog post, tweet) and want to know if it is new: returns New Runtime records that cite that exact URL, related records and patterns, the source's hub, and next steps. Free, no auth. Example: {"url": "https://github.com/modelcontextprotocol/servers", "task": "is this new?"}. |
get_forecastsread-only | Use when you need falsifiable predictions about AI engineering with a track record: New Runtime's dated forecasts (claim, probability, due date, source analysis) and the public scorecard (hit rate, Brier score); resolved ones carry the grade and evidence. Free, no auth. Example: {"status": "open", "limit": 10}. |
get_latest_metadoxread-only | 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. |
get_recordread-only | Use after search_archive, whats_new or what_changed to read one record in full: title, dates, source URLs, topics and body text. Pass the newruntime.com URL you got, or kind + slug. Free, no auth. Example: {"url": "https://newruntime.com/signals/<slug>/"}. |
get_submission_statusread-only | Moderation status of a source, correction or usage receipt you submitted (received, accepted, rejected...), by its as_... submission id. Shows only the public moderation fields. |
get_topic_briefread-only | 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 news signals with URLs. Free, no auth. Example: {"topic": "coding-agents"}. Topic list: https://newruntime.com/topics.json. |
list_agent_tasksread-only | 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"}. |
list_paid_toolsread-only | 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. |
search_archiveread-only | 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. Searches New Runtime's Field Notes, Raw Signals and Analyses; returns up to 10 hits with title, date, one-line summary and URL. Free, no auth. Example: {"query": "MCP security", "limit": 5}. Then get_record for the full text. |
submit_correction | 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. |
submit_source | 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. |
submit_usage_receipt | 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. |
what_changedread-only | 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, which forecasts resolved, and which new items are new facts versus new angles on known stories, plus what still holds. Saves re-reading and re-deduplicating the web. Free, no auth. Example: {"since": "2026-09-01", "topic": "mcp"}. |
whats_newread-only | Use for a quick feed of what New Runtime published in the last 1-14 days (news signals, analyses, field notes, weekly knowledge updates), newest first, at most 40 items with title, date, URL and a one-line why. Free, no auth. Example: {"days": 7, "kinds": ["signal", "analysis"]}. For what changed against what you already knew, use what_changed. |
Schemas: list_tools.
{
"name": "com.newruntime/archive",
"_meta": {
"io.modelcontextprotocol.registry/publisher-provided": {
"prompts": [
"research_a_url",
"weekly_ai_engineering_brief",
"verify_a_claim"
],
"toolsExposed": true,
"readOnlyTools": [
"search_archive",
"get_record",
"list_paid_tools",
"contextualize_url",
"whats_new",
"what_changed",
"get_forecasts",
"list_agent_tasks",
"get_latest_metadox",
"get_topic_brief",
"get_submission_status"
],
"resourcesOnly": false,
"writeToolsExposed": false,
"publicSourcePolicy": "Only published public records and publishable source URLs are exposed. Submissions are queued for weekly review and never change the site directly.",
"moderatedSubmissionTools": [
"submit_source",
"submit_correction",
"submit_usage_receipt"
]
}
},
"title": "New Runtime public records",
"$schema": "https://static.modelcontextprotocol.io/schemas/2025-12-11/server.schema.json",
"remotes": [
{
"url": "https://newruntime.com/mcp",
"type": "streamable-http"
}
],
"version": "0.7.0",
"websiteUrl": "https://newruntime.com/agents/",
"description": "AI-engineering archive: search, what is new, forecasts, topic briefs, URL context; moderated inbox."
}