Bolna API + MCP by Bolna

# Fix list: Bolna API + MCP

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/bolna, the October 2026 research run, assessed 1 October 2026. Grade D, 52.9 out of 100.

This is everything the published grade says the listing lacks, the biggest possible gain to the total first. It comes from the reason given for each score, the checklist each category was scored against (https://www.anchorterminal.com/benchmark/#checklist), the provenance checks, the deductions, what we couldn't check and what the review panel asked for. A fix counts at the next check, once it's public.

For a coding agent working on Bolna API + MCP: work through the items below in the product, its docs and its public pages. Each category gives the reason for its score, with the points each checklist item earned, and the checklist itself, so the gap is the items that earned less than their points. Change the product, not the wording, and keep a note of what you changed and where it's published.

## 1. Reliability, 50 out of 100, up to 10 more on the total

Why it scored 50: A status page sits at status.bolna.ai, but its robots rules blocked our reader, so we couldn't confirm components or history (10 of 20 for the page, 5 for unreadable history). Published limits are 500 requests a minute on `/call` and execution reads, 1,000 a minute elsewhere, 2 concurrent calls on trial accounts and 10 on paid ones (15). The rate-limit page says a 429 comes back and tells callers to use exponential backoff, with no Retry-After header and no idempotency keys for call creation (10 of 15). No SLA found for any tier (0). The API is generally available (10).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-reliability):

Hosted APIs, MCP servers, models and platforms.

- 20, a public status page with component history (Statuspage, Instatus, BetterStack or the vendor's own).
- 0 to 30, the incident record for the last 90 days on that page. 30 for a clean record or trivial incidents only, 20 for minor incidents only, 10 for one major outage (an hour or more of a core API down, or errors across the board), 0 for several. 5 when there's no history we could read, and the note says so.
- 15, rate limits documented with numbers.
- 15, documented 429 or overload handling (Retry-After, backoff guidance), and idempotency keys or safe-retry guidance where writes are involved.
- 10, an SLA published for any paid tier.
- 10, the surface agents use is generally available, not beta or preview.

Local packages, SDKs, frameworks and stdio MCP servers.

- 20, installs from an official package with supported runtimes stated.
- 25, a public CI and test suite, passing on the default branch.
- 0 to 25, open crash or regression issues relative to activity (25 for few and handled, 0 for many, old and unanswered).
- 15, semver discipline and breaking changes called out in a changelog.
- 15, version 1.0 or later, or declared stable.

Protocols are read from their reference implementations, the public facilitators or servers, spec stability and test vectors.

## 2. Security & auth, 44 out of 100, up to 9.8 more on the total

Why it scored 44: Bearer keys shown once and stored hashed, revocable from the dashboard, with separate `sa-` keys for enterprise sub-accounts (25). We take 5 off because every MCP tool accepts an `api_key` argument, which puts a secret into the model's context (20 of 30). Destructive MCP tools are flagged so clients ask first, but there's no read-only key (10 of 20). Callers' speech is untrusted input and we found no prompt-injection guidance. An open GitHub issue (#899, 30 July 2026) reports that the open-source framework's follow-up webhook skips SSRF checks (3 of 15). Each execution keeps a record and raw logs, but we found no audit log of account actions (8 of 15). No security.txt, no bug bounty, and no SOC 2 or ISO 27001 claim. The docs cite an A+ penetration-test rating and send certification questions to support (3 of 20). Webhooks and tool calls carry no signature, only three fixed source IPs.

The checklist (https://www.anchorterminal.com/benchmark/#checklist-security):

- 0 to 30, the credential model. 30 for OAuth 2.1 with scopes, or scoped and revocable keys with rotation. 20 for plain revocable API keys. 10 for one all-powerful key. 10 off when a secret can travel in a URL query string as a documented option.
- 0 to 20, read-only or least-privilege modes, and confirmation or approval for destructive actions.
- 0 to 15, prompt-injection posture where the tool returns untrusted content (documented mitigations or guidance). A tool that returns no untrusted content gets 10.
- 0 to 15, audit logs or per-call visibility for the operator.
- 0 to 20, a security programme. security.txt or a disclosure policy, a bug bounty, SOC 2 or ISO 27001, advisories handled in public.

Models are read for retention, whether API data trains models (and whether that's off by default), zero-retention options and certifications. Frameworks for telemetry defaults, approval hooks, guardrails and sandboxing.

## 3. Payments & pricing, 30 out of 100, up to 8.8 more on the total

Why it scored 30: No x402, MPP or L402 (0 of 40). The $0.06 standard minute and pilot plans are public, but the telephony and platform fees billed on top aren't priced on the pricing page (15 of 20). $5 of free credit at signup, and we found no statement on whether a card is needed (15 of 20). Access starts with a human signup in the dashboard (0).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-payments):

The published rubric, also on the [x402 page](https://www.anchorterminal.com/x402/).

- 40, a machine payment protocol (x402, MPP or L402) on the tool's own endpoints. 10 to 30 when it covers only some endpoints or only goes through a third party, and the note says which.
- 20, per-call or per-unit pricing published without a login. 10 for public plan-only pricing, 0 for "contact sales" or prices behind a login.
- 20, a free tier or trial that doesn't need a card.
- 20, autonomous onboarding, meaning an agent can get access without a person signing up in a browser (keyless use, x402, a programmatic key API).

Payment platforms and agent wallets rarely charge for their own API over a machine protocol, so the first line has steps for them, and the highest one that applies counts. 40 when x402, MPP or L402 runs on all their own endpoints, 30 when it runs on part of their own API, 25 when their merchants can accept one, 20 for running a facilitator, 15 for paying as a buyer, and 0 when the only protocol is their own. Merchant acceptance sits above a facilitator because the platform's own customers can charge agents through it, while a facilitator settles for sellers who wire up the protocol themselves. The counter-argument (a facilitator does more for the protocol as a whole) has a point. Each note says which step applied.

Open-source software you run yourself is scored on its hosted or paid option if it has one. A free, self-hosted package with nothing to buy gets 20, 20 and 20 for the last three lines, and 0 to 40 for the first only if it ships a payment protocol.

## 4. Agent ergonomics, 52 out of 100, up to 7.8 more on the total

Why it scored 52: The hosted MCP server loads 84 tools, grouped by area but with no way to load only some (5 of 25). Execution and batch lists page and filter (14 of 20). Error messages are readable and the docs warn about specific traps, such as `scheduled_at` with a `Z` suffix returning 500 (15 of 20). 23 tools carry `destructiveHint`, but nothing makes call creation idempotent (10 of 20). A call needs only `agent_id` and `recipient_phone_number`. The official packages are the open-source Python framework and a browser Web Call SDK, with no server SDK for the hosted API (8 of 15).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-ergonomics):

- 0 to 25, context cost. For MCP, the number and size of the tool definitions (25 for ten or fewer compact tools, 15 for 11 to 30, 5 for more than 30, plus up to 10 back for toolsets, dynamic loading or read-only subsets). For APIs, whether responses can be sized (field selection, limits, summaries).
- 20, pagination, filtering and output-size controls.
- 20, actionable, documented error responses, codes and messages an agent can recover from.
- 20, idempotency or safe retries, and for MCP the `readOnlyHint` and `destructiveHint` annotations.
- 15, sensible defaults, few required parameters, and official SDKs in at least two languages.

Models are read for tool use, structured output, prompt caching, context length, batch and SDKs. Frameworks for how much code and how many defaults a tool-calling agent with MCP needs.

## 5. Schema & documentation, 63 out of 100, up to 6 more on the total

Why it scored 63: llms.txt links an OpenAPI file at `/docs/api-reference/openapi.yml`, but the link returned 404 on 2026-10-01. The MCP server flags destructive tools with `destructiveHint` in its definitions, though we couldn't read the full input schemas, so 5 of 25. llms.txt plus Markdown copies of every page (10). Endpoint pages describe purpose and the MCP tool list gives one line per tool, with little on when not to use one (12 of 20). Parameters are typed per endpoint, but agent creation takes large nested `agent_config` and `agent_prompts` objects (9 of 15). An errors page documents the `error` and `message` format and the status codes, and endpoint pages carry examples (12 of 15). v2 endpoints, a dated changelog and retirement notices (15).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-schema):

APIs and MCP servers.

- 25, a machine-readable contract (a public OpenAPI file or similar; for MCP, typed JSON Schema inputs on every tool).
- 10, llms.txt or Markdown docs served for agents.
- 0 to 20, descriptions that say what a tool is for, when to use it and when not to, read from the tool definitions in the source or the API reference.
- 0 to 15, typed inputs with enums, constraints and required fields, and no free-form JSON blobs.
- 0 to 15, examples and documented error responses.
- 15, versioning and a public changelog.

Models are read from the API reference, the OpenAPI file, llms.txt, the structured-output and tool-use docs and the model cards. Frameworks from docs a model can follow, typed interfaces, examples and the API reference.

## 6. Transparency & trust, 70 out of 100, up to 2.6 more on the total

Made of editorial 57, provenance 82.

Why it scored 70: The core orchestration framework is MIT on GitHub, while the hosted API, MCP server and dashboard are closed (22 of 30). The privacy policy (28 March 2025) names Whismurwave Inc. and keeps data while the account is active and for up to 3 years of inactivity, the security page says recording retention is set through support, and the terms name Voxlabs Private Limited. No DPA found (10 of 30). The changelog dates its retirements, such as legacy extractions on 18 September 2026 with a migration guide (15 of 20). Data locations are stated (AWS us-east-1 by default, ap-south-1 for India), but we found no subprocessor list (10 of 20).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-transparency):

- 0 to 30, source availability and licence clarity. 30 for open source under an OSI licence, 15 for closed with clear terms, 0 for unclear terms.
- 0 to 30, data handling and retention statements that agree with each other (privacy policy, DPA, retention periods, subprocessors).
- 0 to 20, a deprecation policy or notices with dates.
- 0 to 20, telemetry disclosed with an opt-out (local software), or subprocessors and data locations disclosed (hosted).

The other half of Transparency and trust is the provenance score, computed from checked facts (below). The category score is the mean of the two.

Provenance checks not met in full (half of this category, computed from checked facts):

- Domain age: bolna.ai, registered 2024-09-27 (2 years) (7 of 15)
- security.txt: not found (0 of 10)

## 7. Maintenance & community, 76 out of 100, up to 2.1 more on the total

Why it scored 76: PyPI release 0.10.266 on 2026-09-29 and changelog entries up to 30 September (30). Sixteen dated changelog entries in September alone (20). The open-source repository has 55 open issues and 53 open pull requests, and the most recent issues from late July show no maintainer reply on the page we read (10 of 25). Official packages are the `bolna` Python framework and `@bolna/web-call`, with no server SDK for the hosted API (8 of 15). Python 3.10 or later, daily releases and GitHub Actions in the repository (8 of 10).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-maintenance):

- 0 to 30, time since the last release, or the last published model or API change for a closed service. 30 within 30 days, 20 within 90, 10 within 180, 0 older.
- 20, at least three releases or dated changelog entries in the last 90 days.
- 0 to 25, responsiveness. Issues and pull requests answered on GitHub (the open issues and how recent the replies are). For closed services, a public changelog and a support or community channel that answers, 0 to 15.
- 15, presence in the official MCP registry under a verified namespace (MCP servers), or current official SDKs (APIs and models).
- 10, package health, current dependencies and CI.

Models are read for deprecation notice periods and model churn rather than release counts.

## What we couldn't check

What we couldn't read counted as absent. Publishing it on a page a plain HTTP fetch can read (not only in a browser) lets the next check count it.

- We couldn't read status.bolna.ai, so the incident record for the last 90 days is unknown.
- Whether signup needs a card for the $5 credit.
- The per-minute telephony and platform fees billed on top of the $0.06 rate.
- The listing's toolCount of 91 was wrong. The tool list page says 84 tools, 82 backed by the REST API.

## Weaknesses

- Unsigned webhooks and tool calls, IP allowlisting only
- No security.txt, bug bounty or SOC 2 claim found
- The OpenAPI link in llms.txt returns 404
- Telephony and platform fees aren't priced on the pricing page
- Terms and privacy policy name different legal entities

## What costs an agent a turn today

The notes we give agents before they call it. Each one is a workaround an agent shouldn't need.

- Wait for the `completed` execution status, not `call-disconnected`, before reading cost, recording or extracted data
- Send `scheduled_at` with a numeric offset such as `+00:00`, since a `Z` suffix returns 500
- Back off exponentially on 429, the limit on `/call` is 500 requests a minute
- Reject webhook and tool requests that don't come from 13.203.39.153, 13.126.9.249 or 13.202.133.53
- Leave the MCP `api_key` argument empty unless acting as a sub-account, so the key stays out of the transcript

## What the review panel asked for

- let readers fetch the status page
- add idempotency keys
- HMAC-signed webhooks
- no key argument on MCP tools

## When it's done

Send what changed and where it's published as a dispute (https://www.anchorterminal.com/builders/#disputes, or `POST https://www.anchorterminal.com/api/v1/contact` with `"kind": "dispute"`). Disputes are answered in public, and the listing is checked again by the same checklist. Paying for an audit or a listing claim changes nothing here.