The White Report
Server Details
The White Report news archive: crypto and AI stories, search, full text, ticker, graded calls.
- Status
- Healthy
- Uptime
- 100.0% over the last 24 hours
- Last Tested
- Transport
- Streamable HTTP · MCP 2025-11-25
- URL
- Repository
- CYAN-ECLIPSE-TECHNOLOGIES-LLC/white-report-mcp
- GitHub Stars
- 0
- Server Listing
- white-report
- ChatGPT
- –
- Claude
- –
TDQS
Scored across 12 tools
Each tool targets a distinct resource and access pattern (article by slug vs. newest vs. keyword search; call envelopes vs. human-readable scoreboard; telemetry latest vs. series), and descriptions explicitly cross-reference the 'use X instead' alternative. The only soft overlap is the three-way calls_list / calls_since / get_scoreboard split, which is same data in different shapes and requires reading descriptions to choose correctly.
Three competing conventions coexist: noun_verb (calls_get, calls_list, calls_since, event_brief_get, event_briefs_list), verb_noun (get_article, get_latest, search_articles), and noun_noun (telemetry_latest, telemetry_series). There is also singular/plural drift within one family (event_brief_get vs. event_briefs_list). Each family is internally readable, but the set as a whole has no single predictable pattern.
Twelve tools is well within a healthy range and each maps to a genuine need: three article-access modes, a call envelope trio plus scoreboard, event brief get/list, telemetry latest/series, and a market snapshot. No tool looks redundant or padded for its stated purpose.
The read-only surface covers all four domains (articles, calls, event briefs, telemetry) with both lookup and enumeration on each side, plus a scoreboard and ticker for the publication's core value. Minor gaps remain: no category/date-range article browsing, no documented pagination or offset on search_articles or get_latest beyond a clamped limit of 20, so deep archive traversal is a dead end.
Available Tools
12 toolscalls_getGet One Call (Machine Format)ARead-onlyIdempotentInspect
Machine format (wr.calls.v1, testnet): the full sealed envelope for one call id (e.g. ep.6). Use calls_list first if you do not have a call_id. Returns the full sealed envelope object for that call. An unknown call_id returns an error; if the store file is absent on this deployment, it returns {call: null, note} instead. Behaviour: read-only and public — no authentication; every call is logged in aggregate (tool name, query, result count, coarse client class; no IP addresses).
Inputs: call_id (required).
calls_listList Calls (Machine Format)ARead-onlyIdempotentInspect
Machine format (wr.calls.v1, testnet): sealed per-call envelopes from the graded-calls ledger. Optional filters: state (grade_state), asset (substring of asset_or_event). Historical record only. Use calls_get for one call's full envelope once you have its call_id, calls_since to sync only what is new, and get_scoreboard instead for the human-readable track record. state matches grade_state exactly; asset is a case-insensitive substring of asset_or_event. If the feed is frozen pending reconciliation, this returns frozen:true with count 0 rather than stale data; if the store file is absent on this deployment, it returns count 0 with a note — neither is an error. Returns {format, frozen, ledger_count, count, calls} with full sealed envelopes. Behaviour: read-only and public — no authentication; every call is logged in aggregate (tool name, query, result count, coarse client class; no IP addresses).
Inputs: asset, state.
calls_sinceCalls Since Timestamp (Machine Format)ARead-onlyIdempotentInspect
Machine format (wr.calls.v1, testnet): call envelopes with published_at after the given ISO-8601 UTC timestamp (v1 basis: published_at). Use calls_list instead when you need state/asset filters rather than a time window. The comparison is strict — only envelopes published after the timestamp are returned — and v1 covers publication times only, not later grade changes. Returns {format, since, basis, count, calls}; if the store file is absent on this deployment, count 0 with a note. Behaviour: read-only and public — no authentication; every call is logged in aggregate (tool name, query, result count, coarse client class; no IP addresses).
Inputs: since (required).
event_brief_getGet Event Brief (Machine Format)ARead-onlyIdempotentInspect
Machine format (wr.event.v1, testnet): one sealed event brief envelope by id; version optional (default: latest version). Use event_briefs_list first if you do not have a brief id. Without version, returns the highest-numbered version of the brief; with version, exactly that version. An unknown id or version returns an error; if the store file is absent on this deployment, it returns {brief: null, note}. Returns the full sealed envelope. Behaviour: read-only and public — no authentication; every call is logged in aggregate (tool name, query, result count, coarse client class; no IP addresses).
Inputs: id (required), version.
event_briefs_listList Event Briefs (Machine Format)ARead-onlyIdempotentInspect
Machine format (wr.event.v1, testnet): the scheduled-event coverage calendar and brief index (id, version, event, sla_met). Use event_brief_get for one brief's full sealed envelope once you have its id. Returns the calendar plus a brief index only — id, version, published_at, event, sla_met per brief — not the full envelopes. Takes no parameters. If the store file is absent on this deployment, the calendar and briefs are empty with a note. Behaviour: read-only and public — no authentication; every call is logged in aggregate (tool name, result count, coarse client class; no IP addresses).
Inputs: no named parameters.
get_articleGet ArticleARead-onlyIdempotentInspect
Full machine-readable article (the JSON twin: facts, sources, call, full text). Use search_articles or get_latest first if you do not have a slug. Returns the article's complete JSON twin exactly as published — headline, summary, key facts, sources, the call the article makes (if any), and the full text. The slug must be lowercase letters, digits, and hyphens; an unknown or malformed slug returns an error, never a partial result. Behaviour: read-only and public — no authentication or API key; served from the site's published article files, and every call is logged in aggregate (tool name, query, result count, coarse client class; no IP addresses).
Inputs: slug (required).
get_latestGet Latest ArticlesARead-onlyIdempotentInspect
Newest White Report articles (headline, summary, dates, URLs). Start here for the newest coverage; use search_articles instead when you have a topic or keyword, and get_article for one article's full text once you have its slug. Returns briefs only — headline, url, slug, datePublished, category, summary, tickers, entities, and up to 5 keyFacts per article — newest first; undated legacy articles are excluded (search_articles reaches them). limit is clamped to 1-20. Returns {articles, note}. Behaviour: read-only and public — no authentication or API key; answers are served from the site's published article files, not a live database, and every call is logged in aggregate (tool name, query, result count, coarse client class; no IP addresses).
Inputs: limit.
get_scoreboardGet ScoreboardARead-onlyIdempotentInspect
Every dated public call The White Report has made, its grade date, and its grade (null while pending). Start here for the track record in human-readable form; use calls_list instead for the sealed machine-format envelopes (wr.calls.v1) behind the same calls. Takes no parameters. Returns the scoreboard exactly as published in the site's calls file — the same record the website shows, with grade null on calls still pending grading. Behaviour: read-only and public — no authentication or API key; every call is logged in aggregate (tool name, result count, coarse client class; no IP addresses).
Inputs: no named parameters.
get_tickerGet Market TickerARead-onlyIdempotentInspect
Market snapshot: crypto top 5, SI stocks, and metrics, with the snapshot timestamp. For the publication's dated predictions and their grades, use get_scoreboard instead. This is a snapshot baked when the site was last built, not live prices — the as_of field gives its timestamp, and the live ribbon on the website refreshes more often. Takes no parameters. Returns the snapshot object plus as_of and note fields. Behaviour: read-only and public — no authentication or API key; every call is logged in aggregate (tool name, result count, coarse client class; no IP addresses).
Inputs: no named parameters.
search_articlesSearch ArticlesARead-onlyIdempotentInspect
Search the article archive by keyword (headline, summary, key facts, entities, tickers). Use get_latest instead for the newest articles without a query; open a hit's full text with get_article (the slug comes from the search results). Matching is a case-insensitive substring match of the whole query against each article's headline, summary, key facts, entities, and tickers — not a semantic search; headline matches rank first, then newest first. An empty query returns an error. count is the number of articles returned. Returns {query, count, articles} with the same brief shape as get_latest. Behaviour: read-only and public — no authentication or API key; served from the site's published article files, and every call is logged in aggregate (tool name, query, result count, coarse client class; no IP addresses).
Inputs: query (required), limit.
telemetry_latestLatest Telemetry (Machine Format)ARead-onlyIdempotentInspect
Machine format (wr.telemetry.v1, testnet): the latest sealed compile envelope per telemetry metric; optional metric filter. Use telemetry_series instead for every envelope of one metric over time. The latest point per metric is chosen by its as_of timestamp; stale_flags lists metrics whose newest point the store marks stale. If the store file is absent on this deployment, latest is empty with a note. Returns {format, latest, stale_flags}. Behaviour: read-only and public — no authentication; every call is logged in aggregate (tool name, query, result count, coarse client class; no IP addresses).
Inputs: metric.
telemetry_seriesTelemetry Series (Machine Format)ARead-onlyIdempotentInspect
Machine format (wr.telemetry.v1, testnet): every sealed compile envelope for one metric id, oldest first. Use telemetry_latest instead when you only need the newest envelope per metric. Points are ordered oldest first by as_of. An unknown metric id is not an error — it returns count 0 and empty points; call telemetry_latest with metric omitted to enumerate the valid metric ids. stale says whether this metric is in the store's stale_flags. Returns {format, metric, count, points, stale}; if the store file is absent on this deployment, empty points with a note instead. Behaviour: read-only and public — no authentication; every call is logged in aggregate (tool name, query, result count, coarse client class; no IP addresses).
Inputs: metric (required).
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Related MCP Connectors
AI-curated digital asset news: search, sentiment and quality filters, AI-generated daily digests.
Real-time curated crypto news for AI agents with sentiment, recaps, and search.
Crypto news, perps, market pulse, price history and token checks for AI agents. USDC per call.
Free crypto news for agents: top stories by market impact, search, topics, AI digest.
Related MCP Servers
- AlicenseAqualityCmaintenanceRead-only, source-linked news intelligence for AI agents: search The Neural Ledger's stories, retrieve story details with citations and revision history, and resolve related entities and assets. It is an evidence layer, not a trading or execution service.82MIT
- AlicenseAqualityDmaintenanceProvide the latest cryptocurrency news to AI agents.174MIT
- AlicenseAqualityDmaintenanceNarrative & signal intelligence for AI agents: crypto/AI/macro convergence & divergence.21MIT
- AlicenseAqualityDmaintenanceA cryptocurrency news aggregation server providing AI-powered sentiment ratings, trading signals, and real-time market updates. It allows users to search news by coin, source, or engine type and filter results based on AI impact scores and bullish/bearish signals.112,437MIT
Glama MCP Gateway
Add one secure layer between your agents and this server.