@tensorfeed/mcp-server 1.28.0 → 1.30.0

package/CHANGELOG.md

@@ -2,6 +2,30 @@
 
 All notable changes to the [TensorFeed.ai MCP server](https://github.com/RipperMercs/tensorfeed-mcp). Free tools work without configuration; premium tools require a bearer token via the `TENSORFEED_TOKEN` env var. Buy credits at [tensorfeed.ai/developers/agent-payments](https://tensorfeed.ai/developers/agent-payments).
 
+## 1.30.0 - 2026-05-14
+
+### Changed
+- Every paid tool description now advertises the USD price alongside the credit cost. "Costs 1 credit." -> "Costs 1 credit ($0.02).". Lets agents budget calls without needing to look up credit-to-USD conversion separately. No behavior change; pure description metadata.
+
+### Changed (strict-premium tier)
+- Six tools now flagged as "Strict premium, no free trial" in their descriptions: `pricing_series`, `benchmark_series`, `status_uptime`, `status_leaderboard`, `provider_deepdive`, `probe_series_premium`. These map to TF Worker paths in the strict-premium set (worker/src/strict-premium-endpoints.ts) which bypasses the per-IP 100-call/day free trial. Their free 7-day-capped siblings (`pricing_series_free`, `benchmark_series_free`, `status_uptime_free`, `status_leaderboard_free`) remain the discovery option. Agents that hit these without auth get a 402 immediately with a strict-premium-specific message, not the free-trial advert.
+- Rationale: per the TF 70/30 premium-lock framework, full-window historical time-series and heavy aggregations are TF's moat and warrant strict gating during the demand-discovery phase. Pricing per call is unchanged at this version; repricing into a higher tier is a separate future decision.
+
+## 1.29.0 - 2026-05-13
+
+### Added
+- `get_honeypot_iocs` (free): wraps `/api/security/iocs.json`. First-party honeypot indicators-of-compromise observed against TensorFeed.ai trap endpoints over the last 30 days. Filter by `type` (ip / ua / path), cap with `limit` (default 50, max 500). License CC0; pure first-party observation, not a re-export. Pairs with `get_ai_supply_chain_iocs` (GHSA-derived) for a two-layer defender feed.
+- `check_afta_certification` (free): wraps `/api/afta-certify/check`. Runs 6 deterministic AFTA probes (well-known endpoints, x402 manifest, pricing transparency, free trial, federation membership, signed receipts) against any domain and returns score, verdict, and per-check pass/fail. Useful for federation prospects, agent governance audits, and demand-side trust signals.
+- `get_agent_reputation_card` (free): wraps `/api/agents/reputation/{wallet}` and `/api/agents/reputation/by-token/{prefix}`. Returns the Agent Reputation Bureau card for an EVM wallet or tf_live_ token prefix: reliability, activity, spend, streak, composite trust score, trust grade (S/A/B/C/D), flags, claim status, leaderboard ranks. Cards rebuild daily 04:50 UTC. Returns a helpful 404 hint when no card exists yet.
+
+### Added (MCP primitive)
+- `tensorfeed://news/latest` resource: latest 20 AI news articles as a JSON resource. First TF use of the MCP `resources` primitive, exposed alongside the existing `get_ai_news` tool. Hosts that prefer data-shaped surfaces (Claude Desktop, some agent frameworks) can attach the news feed directly without going through tool-call decision loops.
+
+### Fixed
+- `registerTool` now auto-infers `PREMIUM_READ_TOOL` annotations (`idempotentHint: false`) for any tool whose description includes "Costs N credit". Previously all read tools defaulted to `idempotentHint: true`, which meant agents that retried paid calls on transient failure could double-charge users. Affects every paid read tool (premium_routing, pricing_series, benchmark_series, status_uptime, status_leaderboard, premium_agents_directory, whats_new, compare_models, provider_deepdive, cost_projection, news_search, probe_series_premium). Write tools (create_*_watch / delete_watch) keep their explicit CREATE_TOOL / DELETE_TOOL annotations and bypass the inference.
+
+Tool count: 46 -> 49. First MCP resource live. Annotations now spec-honest about retry semantics.
+
 ## 1.28.0 - 2026-05-13
 
 ### Added

package/dist/index.js

@@ -78,6 +78,16 @@ const READ_TOOL = {
     idempotentHint: true,
     openWorldHint: true,
 };
+// Paid read tools: same as READ_TOOL but idempotentHint: false because each
+// call burns a credit. Agents that retry on transient failure under
+// READ_TOOL semantics would double-charge users on flaky networks. Applied
+// automatically below when a tool's description includes "Costs N credit".
+const PREMIUM_READ_TOOL = {
+    readOnlyHint: true,
+    destructiveHint: false,
+    idempotentHint: false,
+    openWorldHint: true,
+};
 const CREATE_TOOL = {
     readOnlyHint: false,
     destructiveHint: false,
@@ -90,15 +100,50 @@ const DELETE_TOOL = {
     idempotentHint: true,
     openWorldHint: true,
 };
-function registerTool(name, description, paramsSchema, cb, annotations = READ_TOOL) {
+const PAID_TOOL_RE = /Costs \d+ credit/i;
+function registerTool(name, description, paramsSchema, cb, annotations) {
+    // When no explicit annotation is passed, infer from the description.
+    // Free reads: READ_TOOL (idempotent). Paid reads (description mentions
+    // "Costs N credit"): PREMIUM_READ_TOOL (NOT idempotent, because each
+    // call burns a credit and naive retry would double-charge). Tools that
+    // mutate (create_*_watch / delete_watch) pass CREATE_TOOL / DELETE_TOOL
+    // explicitly and bypass this inference.
+    const finalAnnotations = annotations ?? (PAID_TOOL_RE.test(description) ? PREMIUM_READ_TOOL : READ_TOOL);
     const wrapped = (async (args, extra) => {
         const result = await cb(args, extra);
         return sanitizeToolResponse(result);
     });
     // Direct dispatch to the SDK's underlying registration. Do not call
     // registerTool() here or it will recurse forever.
-    return server.tool(name, description, paramsSchema, annotations, wrapped);
+    return server.tool(name, description, paramsSchema, finalAnnotations, wrapped);
 }
+// ── Resource: tensorfeed://news (latest 20 AI articles) ─────────────
+// MCP's `resources` primitive is a parallel surface to tools. Hosts can
+// list resources, subscribe to them, and fetch them by URI without going
+// through the tool-call decision loop. We expose the latest news as a
+// resource AS WELL AS the get_ai_news tool, so clients that prefer
+// data-shaped surfaces (Claude Desktop, some agent frameworks) can attach
+// the news feed directly. Same upstream endpoint; different MCP primitive.
+server.registerResource('latest_news', 'tensorfeed://news/latest', {
+    description: 'Latest 20 AI news articles aggregated by TensorFeed.ai. Refreshed every ~10 minutes from 15+ sources (Anthropic, OpenAI, Google, TechCrunch, The Verge, arXiv, and more). Each row carries title, source, URL, snippet, categories, and publication timestamp. JSON payload. Refresh on `resources/read`; no subscription required.',
+    mimeType: 'application/json',
+}, async (uri) => {
+    const data = (await fetchJSON('/news?limit=20'));
+    return {
+        contents: [
+            {
+                uri: uri.toString(),
+                mimeType: 'application/json',
+                text: JSON.stringify({
+                    source: 'TensorFeed.ai',
+                    fetched_at: new Date().toISOString(),
+                    count: data.articles.length,
+                    articles: data.articles,
+                }, null, 2),
+            },
+        ],
+    };
+});
 // ── Tool: get_ai_news ───────────────────────────────────────────────
 registerTool('get_ai_news', 'Get the latest AI news articles from TensorFeed.ai. Aggregates from 15+ sources including Anthropic, OpenAI, Google, TechCrunch, The Verge, arXiv, and more.', {
     category: z.string().optional().describe('Filter by category (e.g. "anthropic", "openai", "research", "tools")'),
@@ -323,6 +368,114 @@ registerTool('get_ai_supply_chain_iocs', 'Get the latest AI/ML supply-chain indi
     const footer = `\n\nPosture: ${data.posture}`;
     return { content: [{ type: 'text', text: `${header}\n\n${body}${footer}` }] };
 });
+// ── Tool: get_honeypot_iocs ─────────────────────────────────────────
+registerTool('get_honeypot_iocs', 'Get the TensorFeed honeypot indicators-of-compromise feed. Returns attacker IP addresses, user agents, and target paths observed against TensorFeed.ai trap endpoints over the last 30 days. Pure first-party observation; not a re-export of upstream threat intel. License CC0; downstream defenders may ingest and pre-block at their edge. Each row: ioc value, ioc type (ip / ua / path), first_seen, last_seen, hit_count. Useful as a low-noise honeypot signal layered on top of public IOC feeds.', {
+    type: z
+        .enum(['ip', 'ua', 'path'])
+        .optional()
+        .describe('Filter to one IOC type. Omit to return all types.'),
+    limit: z
+        .number()
+        .min(1)
+        .max(500)
+        .optional()
+        .describe('Maximum number of IOCs to return (default 50, max 500)'),
+}, async ({ type, limit }) => {
+    const data = (await fetchJSON('/security/iocs.json'));
+    const typeWanted = type?.toLowerCase();
+    const filtered = data.iocs.filter((e) => !typeWanted || e.type.toLowerCase() === typeWanted);
+    const cap = Math.min(limit ?? 50, 500);
+    const rows = filtered.slice(0, cap);
+    if (rows.length === 0) {
+        return {
+            content: [
+                {
+                    type: 'text',
+                    text: `No honeypot IOCs matched (total in feed: ${data.total_iocs}, window ${data.window_hours}h, generated ${data.generated_at}).`,
+                },
+            ],
+        };
+    }
+    const body = rows
+        .map((e, i) => `${i + 1}. [${e.type}] ${e.ioc}\n   hits: ${e.hit_count}  first: ${e.first_seen}  last: ${e.last_seen}`)
+        .join('\n');
+    const header = `TensorFeed honeypot IOCs (${rows.length} of ${filtered.length} matched, ${data.total_iocs} in feed, ${data.window_hours}h window, generated ${data.generated_at}):`;
+    const footer = `\n\nLicense: ${data.policy.license}. ${data.policy.description}`;
+    return { content: [{ type: 'text', text: `${header}\n\n${body}${footer}` }] };
+});
+// ── Tool: check_afta_certification ──────────────────────────────────
+registerTool('check_afta_certification', 'Check whether a domain is AFTA-certified (Agent Fair-Trade Agreement). AFTA is an open standard: cryptographically-signed receipts, transparent pricing, no-charge guarantees for failed responses, on-chain settlement, and federated trust between participating sites. The check runs 6 deterministic probes against the target domain (e.g. tensorfeed.ai, terminalfeed.io): well-known endpoint, receipts endpoint, x402 payment manifest, pricing transparency, free trial availability, federation membership. Returns score, verdict, and which checks passed/failed.', {
+    domain: z
+        .string()
+        .describe('Domain to certify (e.g. "tensorfeed.ai", "terminalfeed.io"). Bare host, no protocol.'),
+}, async ({ domain }) => {
+    const data = (await fetchJSON(`/afta-certify/check?domain=${encodeURIComponent(domain)}`));
+    const lines = data.checks.map((c) => `  ${c.passed ? 'PASS' : 'FAIL'}  ${c.name}${c.details ? ` (${c.details})` : ''}`);
+    const certified = data.afta_certified ? 'AFTA-CERTIFIED' : 'NOT certified';
+    const next = data.next_step ? `\nNext step: ${data.next_step}` : '';
+    return {
+        content: [
+            {
+                type: 'text',
+                text: `Domain: ${data.domain}\nScore: ${data.score}/${data.max}\nVerdict: ${data.verdict} (${certified})\nChecked: ${data.checked_at}\n\nChecks:\n${lines.join('\n')}${next}`,
+            },
+        ],
+    };
+});
+// ── Tool: get_agent_reputation_card ─────────────────────────────────
+registerTool('get_agent_reputation_card', 'Get a TensorFeed Agent Reputation Bureau card for an EVM wallet address or a tf_live_ token prefix. Cards rebuild daily at 04:50 UTC from on-chain receipts plus TF telemetry. Includes reliability score, paid-call activity, total USDC spent, longest paid streak, composite trust score (0-1), trust grade (S/A/B/C/D), flags (sybil_risk, abuse, paid_free_ratio_low), operator-claim status (wallet -> display name signed by EOA), and per-metric leaderboard ranks. Returns null/404 for unknown agents (TF only builds cards for wallets observed in production). Useful for assessing third-party agent trustworthiness before delegating high-value work.', {
+    wallet: z
+        .string()
+        .optional()
+        .describe('EVM wallet address (0x... 42 chars). Provide either this OR token_prefix.'),
+    token_prefix: z
+        .string()
+        .optional()
+        .describe('First N chars of an agent\'s tf_live_ bearer token. Useful when the agent has not signed an operator-claim yet.'),
+}, async ({ wallet, token_prefix }) => {
+    if (!wallet && !token_prefix) {
+        return {
+            content: [
+                {
+                    type: 'text',
+                    text: 'Provide either `wallet` (0x... EVM address) or `token_prefix` (tf_live_... token prefix). Both are accepted, one is required.',
+                },
+            ],
+        };
+    }
+    const path = wallet
+        ? `/agents/reputation/${encodeURIComponent(wallet)}`
+        : `/agents/reputation/by-token/${encodeURIComponent(token_prefix)}`;
+    let data;
+    try {
+        data = await fetchJSON(path);
+    }
+    catch (e) {
+        const msg = e instanceof Error ? e.message : String(e);
+        if (msg.includes('404') || msg.toLowerCase().includes('not_found')) {
+            return {
+                content: [
+                    {
+                        type: 'text',
+                        text: `No reputation card for ${wallet ?? token_prefix}. TF only builds cards for wallets observed in production; if this is a real agent that has paid TF before, it should appear in the next daily rebuild (04:50 UTC).`,
+                    },
+                ],
+            };
+        }
+        throw e;
+    }
+    // The card schema is rich; render the highlights deterministically and
+    // include the raw JSON so the agent can reason over any field it needs.
+    const card = data.card ?? data;
+    return {
+        content: [
+            {
+                type: 'text',
+                text: `Agent reputation card for ${wallet ?? `token_prefix=${token_prefix}`}:\n\n${JSON.stringify(card, null, 2)}`,
+            },
+        ],
+    };
+});
 // ════════════════════════════════════════════════════════════════════
 // PREMIUM TOOLS (require TENSORFEED_TOKEN env var, paid in USDC on Base)
 // ════════════════════════════════════════════════════════════════════
@@ -358,7 +511,7 @@ registerTool('get_account_usage', 'Show per-endpoint usage for the configured Te
     };
 });
 // ── Tool: premium_routing (1 credit) ────────────────────────────────
-registerTool('premium_routing', 'Get a ranked list of recommended AI models for a task with full score breakdown (quality, availability, cost, latency). Costs 1 credit.', {
+registerTool('premium_routing', 'Get a ranked list of recommended AI models for a task with full score breakdown (quality, availability, cost, latency). Costs 1 credit ($0.02).', {
     task: z.enum(['code', 'reasoning', 'creative', 'general']).optional().describe('Task type the model needs to be good at (default: general)'),
     budget: z.number().optional().describe('Max blended USD per 1M tokens'),
     min_quality: z.number().min(0).max(1).optional().describe('Minimum quality score in [0, 1]'),
@@ -475,7 +628,7 @@ registerTool('status_leaderboard_free', 'Cross-provider uptime leaderboard, rank
     };
 });
 // ── Tool: pricing_series (1 credit) ─────────────────────────────────
-registerTool('pricing_series', 'Daily price points for one AI model with min/max/delta summary. Default range = last 30 days, max 90 days. Costs 1 credit.', {
+registerTool('pricing_series', 'Daily price points for one AI model with min/max/delta summary. Default range = last 30 days, max 90 days. Costs 1 credit ($0.02). Strict premium, no free trial; the 7-day-capped sibling pricing_series_free is the discovery option.', {
     model: z.string().describe('Model id or display name (e.g. "Claude Opus 4.7" or "claude-opus-4-7")'),
     from: z.string().optional().describe('Start date YYYY-MM-DD UTC (default: 30 days ago)'),
     to: z.string().optional().describe('End date YYYY-MM-DD UTC (default: today)'),
@@ -500,7 +653,7 @@ registerTool('pricing_series', 'Daily price points for one AI model with min/max
     };
 });
 // ── Tool: benchmark_series (1 credit) ───────────────────────────────
-registerTool('benchmark_series', 'Score evolution for a single benchmark on one AI model. Costs 1 credit. Benchmark keys: swe_bench, mmlu_pro, gpqa_diamond, math, human_eval.', {
+registerTool('benchmark_series', 'Score evolution for a single benchmark on one AI model. Costs 1 credit ($0.02). Strict premium, no free trial; the 7-day-capped sibling benchmark_series_free is the discovery option. Benchmark keys: swe_bench, mmlu_pro, gpqa_diamond, math, human_eval.', {
     model: z.string().describe('Model id or display name'),
     benchmark: z.string().describe('Benchmark key (e.g. swe_bench, mmlu_pro, gpqa_diamond, math, human_eval)'),
     from: z.string().optional().describe('Start date YYYY-MM-DD UTC (default: 30 days ago)'),
@@ -526,7 +679,7 @@ registerTool('benchmark_series', 'Score evolution for a single benchmark on one
     };
 });
 // ── Tool: status_uptime (1 credit) ──────────────────────────────────
-registerTool('status_uptime', 'Daily uptime rollup for one provider with operational/degraded/down day counts and uptime % (degraded counts as half-credit). Costs 1 credit.', {
+registerTool('status_uptime', 'Daily uptime rollup for one provider with operational/degraded/down day counts and uptime % (degraded counts as half-credit). Costs 1 credit ($0.02). Strict premium, no free trial; the 7-day-capped sibling status_uptime_free is the discovery option.', {
     provider: z.string().describe('Provider name (e.g. anthropic, openai, google)'),
     from: z.string().optional().describe('Start date YYYY-MM-DD UTC (default: 30 days ago)'),
     to: z.string().optional().describe('End date YYYY-MM-DD UTC (default: today)'),
@@ -553,7 +706,7 @@ registerTool('status_uptime', 'Daily uptime rollup for one provider with operati
     };
 });
 // ── Tool: status_leaderboard (1 credit) ─────────────────────────────
-registerTool('status_leaderboard', 'Cross-provider uptime leaderboard for a custom date range up to 90 days. Same minute-resolution counter source as status_leaderboard_free, but adds incident_count and mttr_minutes (mean time to recover) per provider. Sorted by uptime % DESC. Costs 1 credit.', {
+registerTool('status_leaderboard', 'Cross-provider uptime leaderboard for a custom date range up to 90 days. Same minute-resolution counter source as status_leaderboard_free, but adds incident_count and mttr_minutes (mean time to recover) per provider. Sorted by uptime % DESC. Costs 1 credit ($0.02). Strict premium, no free trial; the 7-day-capped sibling status_leaderboard_free is the discovery option.', {
     from: z.string().optional().describe('Start date YYYY-MM-DD UTC (default: 30 days ago)'),
     to: z.string().optional().describe('End date YYYY-MM-DD UTC (default: today)'),
 }, async ({ from, to }) => {
@@ -579,7 +732,7 @@ registerTool('status_leaderboard', 'Cross-provider uptime leaderboard for a cust
     };
 });
 // ── Tool: premium_agents_directory (1 credit) ───────────────────────
-registerTool('premium_agents_directory', 'Enriched AI agents catalog joined with live status, recent news, agent traffic, flagship pricing, and a 0-100 trending_score. Costs 1 credit.', {
+registerTool('premium_agents_directory', 'Enriched AI agents catalog joined with live status, recent news, agent traffic, flagship pricing, and a 0-100 trending_score. Costs 1 credit ($0.02).', {
     category: z.string().optional().describe('coding, research, general, creative, frameworks'),
     status: z.enum(['operational', 'degraded', 'down', 'unknown']).optional(),
     open_source: z.boolean().optional(),
@@ -612,7 +765,7 @@ registerTool('premium_agents_directory', 'Enriched AI agents catalog joined with
     };
 });
 // ── Tool: whats_new (1 credit) ──────────────────────────────────────
-registerTool('whats_new', 'Agent morning brief: pricing changes, new/removed models, status incidents, and top news from the last 1-7 days. The tool to call when your agent boots up. Costs 1 credit.', {
+registerTool('whats_new', 'Agent morning brief: pricing changes, new/removed models, status incidents, and top news from the last 1-7 days. The tool to call when your agent boots up. Costs 1 credit ($0.02).', {
     days: z.number().min(1).max(7).optional().describe('Window length in days (default 1)'),
     news_limit: z.number().min(1).max(25).optional().describe('Max news headlines (default 10)'),
 }, async ({ days, news_limit }) => {
@@ -648,7 +801,7 @@ registerTool('whats_new', 'Agent morning brief: pricing changes, new/removed mod
     };
 });
 // ── Tool: compare_models (1 credit) ─────────────────────────────────
-registerTool('compare_models', 'Side-by-side comparison of 2-5 AI models. Returns pricing, benchmarks (normalized to a union of keys with null for missing scores), provider status, and recent news per model, plus rankings (cheapest blended, most context, per-benchmark leaderboard). Costs 1 credit.', {
+registerTool('compare_models', 'Side-by-side comparison of 2-5 AI models. Returns pricing, benchmarks (normalized to a union of keys with null for missing scores), provider status, and recent news per model, plus rankings (cheapest blended, most context, per-benchmark leaderboard). Costs 1 credit ($0.02).', {
     ids: z.string().describe('Comma-separated list of 2-5 model ids or display names. Examples: "Claude Opus 4.7,GPT-5.5,Gemini 3" or "opus-4-7,gpt-5-5"'),
 }, async ({ ids }) => {
     const data = (await fetchJSON(`/premium/compare/models?ids=${encodeURIComponent(ids)}`, { auth: true }));
@@ -675,7 +828,7 @@ registerTool('compare_models', 'Side-by-side comparison of 2-5 AI models. Return
     };
 });
 // ── Tool: provider_deepdive (1 credit) ──────────────────────────────
-registerTool('provider_deepdive', 'Everything about an AI provider in one call: live status, all models with pricing + tier + benchmark scores joined in, recent news mentions, and agent traffic. Costs 1 credit. Aggregation IS the value; doing this from free endpoints would take 4 round-trips.', {
+registerTool('provider_deepdive', 'Everything about an AI provider in one call: live status, all models with pricing + tier + benchmark scores joined in, recent news mentions, and agent traffic. Costs 1 credit ($0.02). Strict premium, no free trial. Aggregation IS the value; doing this from free endpoints would take 4 round-trips.', {
     provider: z.string().describe('Provider id or display name (case-insensitive). Examples: anthropic, openai, google, mistral, cohere'),
 }, async ({ provider }) => {
     const data = (await fetchJSON(`/premium/providers/${encodeURIComponent(provider)}`, { auth: true }));
@@ -705,7 +858,7 @@ registerTool('provider_deepdive', 'Everything about an AI provider in one call:
     };
 });
 // ── Tool: cost_projection (1 credit) ────────────────────────────────
-registerTool('cost_projection', 'Project the cost of a token-usage workload across 1-10 AI models. Returns daily/weekly/monthly/yearly totals per model and a ranking by cheapest monthly. Costs 1 credit.', {
+registerTool('cost_projection', 'Project the cost of a token-usage workload across 1-10 AI models. Returns daily/weekly/monthly/yearly totals per model and a ranking by cheapest monthly. Costs 1 credit ($0.02).', {
     models: z.string().describe('One model or comma-separated list of up to 10 (e.g. "Claude Opus 4.7,GPT-5.5,Gemini 3"). Names or ids both work.'),
     input_tokens_per_day: z.number().min(0).describe('Expected daily input token volume'),
     output_tokens_per_day: z.number().min(0).describe('Expected daily output token volume'),
@@ -740,7 +893,7 @@ registerTool('cost_projection', 'Project the cost of a token-usage workload acro
     };
 });
 // ── Tool: news_search (1 credit) ────────────────────────────────────
-registerTool('news_search', 'Full-text search over the TensorFeed news article corpus with optional date range, provider, and category filters. Relevance scoring with recency boost. Costs 1 credit.', {
+registerTool('news_search', 'Full-text search over the TensorFeed news article corpus with optional date range, provider, and category filters. Relevance scoring with recency boost. Costs 1 credit ($0.02).', {
     q: z.string().optional().describe('Free-text query, e.g. "claude opus pricing". Omit to browse latest filtered articles.'),
     from: z.string().optional().describe('Start date YYYY-MM-DD UTC (inclusive)'),
     to: z.string().optional().describe('End date YYYY-MM-DD UTC (inclusive end-of-day)'),
@@ -794,7 +947,7 @@ registerTool('list_watches', 'List the active webhook watches owned by the confi
     return { content: [{ type: 'text', text: `${data.count} active watches:\n\n${rows}` }] };
 });
 // ── Tool: create_price_watch (1 credit) ─────────────────────────────
-registerTool('create_price_watch', 'Register a webhook watch on a model price change. Costs 1 credit. Watch lives 90 days. Each fire is an HMAC-signed POST to callback_url.', {
+registerTool('create_price_watch', 'Register a webhook watch on a model price change. Costs 1 credit ($0.02). Watch lives 90 days. Each fire is an HMAC-signed POST to callback_url.', {
     model: z.string().describe('Model name (e.g. "Claude Opus 4.7")'),
     field: z.enum(['inputPrice', 'outputPrice', 'blended']).describe('Which price field to watch'),
     op: z.enum(['lt', 'gt', 'changes']).describe('Trigger: lt = below threshold, gt = above, changes = any change'),
@@ -819,7 +972,7 @@ registerTool('create_price_watch', 'Register a webhook watch on a model price ch
     };
 }, CREATE_TOOL);
 // ── Tool: create_status_watch (1 credit) ────────────────────────────
-registerTool('create_status_watch', 'Register a webhook watch on a service status transition (e.g. anthropic becomes down). Costs 1 credit. Watch lives 90 days.', {
+registerTool('create_status_watch', 'Register a webhook watch on a service status transition (e.g. anthropic becomes down). Costs 1 credit ($0.02). Watch lives 90 days.', {
     provider: z.string().describe('Provider name (e.g. anthropic, openai)'),
     op: z.enum(['becomes', 'changes']).describe('becomes = transitions to a specific value; changes = any transition'),
     value: z.enum(['operational', 'degraded', 'down']).optional().describe('Required when op is becomes'),
@@ -843,7 +996,7 @@ registerTool('create_status_watch', 'Register a webhook watch on a service statu
     };
 }, CREATE_TOOL);
 // ── Tool: create_digest_watch (1 credit) ────────────────────────────
-registerTool('create_digest_watch', 'Register a scheduled digest webhook that fires daily or weekly with a curated summary of pricing changes (regardless of whether anything dramatic happened). Costs 1 credit. Watch lives 90 days. Set-and-forget for agents that want a periodic snapshot without subscribing to realtime transitions.', {
+registerTool('create_digest_watch', 'Register a scheduled digest webhook that fires daily or weekly with a curated summary of pricing changes (regardless of whether anything dramatic happened). Costs 1 credit ($0.02). Watch lives 90 days. Set-and-forget for agents that want a periodic snapshot without subscribing to realtime transitions.', {
     cadence: z.enum(['daily', 'weekly']).describe('How often the digest fires'),
     callback_url: z.string().describe('HTTPS URL to POST to when the digest fires'),
     secret: z.string().optional().describe('Optional HMAC shared secret'),
@@ -865,7 +1018,7 @@ registerTool('create_digest_watch', 'Register a scheduled digest webhook that fi
     };
 }, CREATE_TOOL);
 // ── Tool: create_leaderboard_rank_watch (1 credit) ──────────────────
-registerTool('create_leaderboard_rank_watch', 'Register a webhook that fires when a provider crosses a rank threshold on the cross-provider 7-day uptime leaderboard. Rank 1 = best (highest uptime). drops_below: was rank<=N, now rank>N (got worse). rises_above: was rank>=N, now rank<N (got better). changes: any rank movement. Costs 1 credit at registration. Watch lives 90 days.', {
+registerTool('create_leaderboard_rank_watch', 'Register a webhook that fires when a provider crosses a rank threshold on the cross-provider 7-day uptime leaderboard. Rank 1 = best (highest uptime). drops_below: was rank<=N, now rank>N (got worse). rises_above: was rank>=N, now rank<N (got better). changes: any rank movement. Costs 1 credit ($0.02) at registration. Watch lives 90 days.', {
     provider: z.string().describe('Provider name or slug (case-insensitive). e.g. claude, openai, gemini, bedrock, azure'),
     op: z.enum(['drops_below', 'rises_above', 'changes']).describe('Trigger condition'),
     threshold: z.number().int().min(1).optional().describe('Required for drops_below / rises_above. Integer rank position.'),
@@ -925,7 +1078,7 @@ registerTool('probe_latest', 'Last 24 hours of measured LLM endpoint latency and
     };
 });
 // ── Tool: probe_series (1 credit) ───────────────────────────────────
-registerTool('probe_series', 'Daily SLA series for one LLM provider, measured by TensorFeed. Returns per-day count, success rate, ttfb p50/p95/p99, total p50/p95/p99, and incident-hour count across the requested window. Provider status pages are politically managed; this is the measured truth. Pairs naturally with premium_routing for picking a model whose SLA you can verify. Costs 1 credit.', {
+registerTool('probe_series', 'Daily SLA series for one LLM provider, measured by TensorFeed. Returns per-day count, success rate, ttfb p50/p95/p99, total p50/p95/p99, and incident-hour count across the requested window. Provider status pages are politically managed; this is the measured truth. Pairs naturally with premium_routing for picking a model whose SLA you can verify. Costs 1 credit ($0.02). Strict premium, no free trial.', {
     provider: z.enum(['anthropic', 'openai', 'google', 'mistral', 'cohere']).describe('LLM provider key'),
     from: z.string().optional().describe('Inclusive start YYYY-MM-DD (default: 30 days before to)'),
     to: z.string().optional().describe('Inclusive end YYYY-MM-DD (default: today UTC)'),
@@ -987,7 +1140,7 @@ registerTool('mcp_registry_snapshot', 'Today\'s summary of the official Model Co
     };
 });
 // ── Tool: mcp_registry_series (1 credit) ────────────────────────────
-registerTool('mcp_registry_series', 'Multi-day series of MCP server registry growth and churn. Returns per-day total servers, active count, and daily added/removed counts across the requested window. The registry itself is open data, but a 30/90-day trend requires daily capture started weeks ago, which TensorFeed has been running. Costs 1 credit.', {
+registerTool('mcp_registry_series', 'Multi-day series of MCP server registry growth and churn. Returns per-day total servers, active count, and daily added/removed counts across the requested window. The registry itself is open data, but a 30/90-day trend requires daily capture started weeks ago, which TensorFeed has been running. Costs 1 credit ($0.02).', {
     from: z.string().optional().describe('Inclusive start date YYYY-MM-DD (default: 30 days before to)'),
     to: z.string().optional().describe('Inclusive end date YYYY-MM-DD (default: today UTC)'),
 }, async ({ from, to }) => {

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@tensorfeed/mcp-server",
-  "version": "1.28.0",
+  "version": "1.30.0",
   "mcpName": "ai.tensorfeed/mcp-server",
   "description": "MCP server for TensorFeed.ai - AI news, status, model data, and premium endpoints (routing, cost projection, forecast, history series, news search, webhook watches with digest tier, enriched directory) for AI agents",
   "main": "dist/index.js",