Changelog

v0.1.8

2026-09-30

• Bug fixes — minor fixes and stability improvements.

v0.1.7

2026-08-21

• X-Request-Id on every response — the header was stamped only on a successful POST /agent/ask, so a 401, 403, 422, 429, or a GET /health call came back without one — exactly the responses you need to report. Every response now carries an id, and a replayed Idempotency-Key returns the original request's id rather than a new one.
• Every error response now has a documented schema — error bodies come in three shapes, not one: coded ({"detail": {"code", "message"}}), plain ({"detail": "..."}), and the list-shaped 422. All three are now in the API reference, and the Errors guide tables every code with the status it arrives on and what clears it.
• New Idempotency guide — covers the 24-hour replay window, the Idempotent-Replayed header, why a replay is never billed twice, and the cases that don't replay.
• Quota and rate-limit headers documented — X-RateLimit-* are now declared on the success response, not just under 429, so there's a reason to read them before you're throttled. The X-Quota-* headers reporting your remaining plan budget are now covered in Billing.
• mode=fast and mode=smart migrated to a current-generation model — the model previously serving both modes has been retired by its provider. Both now run on its successor, and both now fall back to an independent second provider, so an upstream incident no longer surfaces as a 5xx. mode=max runs on a newer primary model with a three-deep chain behind it.
• mode=max completes reliably on long multi-turn sessions — a large message_history could previously outgrow the model's context window and surface an upstream 400 mid-run. The history you submit is now authoritative and is trimmed locally ahead of each step, so extended conversations run to completion instead of failing partway through.
• Deeper on-chain research — POST /agent/ask can now screen a token for scam flags, mint/freeze authority and top-holder concentration, read buy/sell pressure across 5m-24h windows, rank a token's most profitable traders, and surface newly launched tokens filtered by age and liquidity.

v0.1.6

2026-07-14

• Call the agent over MCP — NeuroAPI now speaks the Model Context Protocol. Point any MCP client at https://api.neurobro.ai/api/v1/mcp (remote, Streamable HTTP) with your X-API-Key, and the ask_neuroapi tool runs the same single-turn research as POST /agent/ask — same rate limits, metering, and billing. See the MCP guide to connect.

v0.1.5

2026-06-28

• Trade-setup confidence is now calibrated — when POST /agent/ask includes a trade setup, its 1-10 confidence is anchored to the setup's own risk/reward and structure instead of being a free-form guess. The same setup now scores consistently, and a high number can't sit on a weak risk/reward — so you can threshold on it reliably.
• More reliable tool use on mode=max — improved how mode=max selects and calls tools, so requests that need live data — market metrics, technical analysis, trade setups — complete more consistently instead of occasionally stalling on tool selection.
• Searchable docs — the documentation site now has full-text search; press ⌘K (or Ctrl-K) or use the search bar to jump to any endpoint, guide, or field.

v0.1.4

2026-06-03

• Structured output with a JSON Schema — POST /agent/ask now accepts an optional output_schema (a JSON Schema with an object root) and returns the validated object on a new output field; answer is then null, so exactly one of the two is ever set and you can consume the result as typed data instead of parsing prose. It's non-streaming only: combining it with stream=true returns 400, and a schema that can't drive structured output returns 422 — both before any usage is charged. Conformance is strict on mode=max and best-effort with automatic retries on fast/smart.
• Steer the agent with a system prompt — a new optional system_prompt shapes the agent's tone, role, or output format (e.g. "answer as a fixed-income analyst, in bullet points"). It layers on top of the built-in instructions and can't override the agent's persona or safety guardrails.
• International equity coverage — the agent can now pull quotes, fundamentals, financials, and technicals for stocks listed in Indonesia, Brazil, and Vietnam, and resolve a company name to its ticker — alongside the existing US-equity tools.
• Streaming restored on fast and smart — stream=true on these modes again streams the answer chunk-by-chunk instead of returning it as a single block at the end.
• Lower latency on repeat and social-data requests — multi-turn requests that resend message_history keep more of the response cache warm on fast/smart, and requests that read social-sentiment data no longer stall when that upstream is briefly degraded.

v0.1.3

2026-05-26

• Failed streaming runs are no longer billed — POST /agent/ask with stream=true previously charged you for runs that errored after the response started, while stream=false was already free on the same failure. Both modes now refund, so you only pay for runs that finish cleanly.
• mode=max is sturdier across long conversations — sending a message_history that contained a tool call without its paired result could trip an upstream 400 and abort the run; the agent now repairs the pairing before each step, so multi-turn max sessions finish instead of failing late.
• Proactive, cost-aware trade setups — when your prompt expresses any directional view on a crypto, stock, or commodity, the response now includes a structured trading setup (entry, stop, take-profits, R:R) alongside the prose; setups whose allocation-weighted R:R doesn't clear an assumed fee + slippage cost are rejected outright rather than returned as marginal trades.

v0.1.2

2026-05-23

• Usage alerts — you now receive an email the first time your request usage reaches 80% of your plan's quota for the billing period. It shows how many requests you've used, how many remain, and when the quota resets, so a busy day doesn't end in unexpected 429s. Manage notifications from your dashboard.

v0.1.1

2026-05-16

• Request size limits on POST /agent/ask — requests are now validated against published limits: prompt up to 32,000 characters, and message_history up to 50 messages with each message's content up to 32,000 characters. Oversized requests are rejected with 422 before any quota is spent. The limits travel in the OpenAPI schema.

v0.1.0

2026-05-15

• Initial public surface: GET /health, POST /agent/ask.