Lens (Thought Industries)

# Fix list: Lens (Thought Industries)

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/thought-industries-lens, the October 2026 research run, assessed 21 October 2026. Grade E, 43.3 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 Lens (Thought Industries): 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, 30 out of 100, up to 14 more on the total

Why it scored 30: Read with the hosted lines. status.thoughtindustries.com (Atlassian Statuspage) has component history and added a Lens Knowledge Platform component on 20 October 2026 (20). That component is one day old with no incidents, so there's no 90-day record for Lens. The two incidents in the 90 days before, on 29 July and 26 August, were minor and on the Academy platform (5, as for no readable history). Monthly interaction quotas are published (100, 200, 400 and 500 by plan), but no request rate limit or behaviour at the cap was found (5 of 15). No 429, Retry-After or retry guidance (0). No SLA, and the Lens AI Evaluation Agreement gives the service as is, without warranty (0). The launch calls Lens beta (0). Total 30.

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. Agent ergonomics, 37 out of 100, up to 10.2 more on the total

Why it scored 37: Read for the REST API, since the MCP tool list wasn't readable. /v1/ask takes one required question and returns an answer, a sessionId and source excerpts with titles and URLs, with no way to trim the excerpts (12 of 25). Reporting takes limit (1 to 100), offset, date range, channel, widget and key filters and sorting. The source and document lists show no paging parameters (12 of 20). No error responses documented. The observed 401s carry a code and a message (5 of 20). No idempotency keys or retry guidance, and no MCP annotations could be read (0). One required field to ask, an optional sessionId for follow-ups and a set-up script for five MCP clients, with no official SDKs (8 of 15). Total 37.

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.

## 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). Plan prices are public with interaction quotas ($0, $499 and $999 a month, Enterprise through sales), with no per-interaction price or overage rate (10 of 20). The Free plan has no stated time limit on the pricing pages, and the Quick Start asks for a website, a work email, a password and an email code, with no card. Sign-up wasn't tried (20). A person signs up in a browser and an Admin creates the key (0). Total 30.

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. Security & auth, 56 out of 100, up to 7.7 more on the total

Why it scored 56: Bearer API keys made by Admins in the console. Each key is for one channel (MCP or API), limited to chosen source groups, can expire on a set date, shows its last use and revokes at once. The OpenAPI file names knowledge:query, sources:manage and reporting:read permissions, though the console's key form shows only channel, source groups and expiry. No OAuth (25 of 30). Source groups limit what each key, widget or MCP connection can answer from, and only Admins create keys. Keys with sources:manage can delete sources with no confirmation step documented (12 of 20). Lens returns excerpts of crawled public pages and uploaded files, and no prompt-injection guidance was found (3 of 15). Reporting lists every interaction with its channel, sources and key, and can be filtered by key (10 of 15). No security.txt on any Lens host. Thought Industries' security page names SOC 2 Type II, ISO 27001, PCI DSS, TX-RAMP and HIPAA reports and yearly penetration tests for its Academy platform, while the Lens sub-processors (Vercel, Neon, Clerk, Inngest) are a different stack and no report is stated to cover Lens (6 of 20). Total 56.

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.

## 5. Schema & documentation, 56 out of 100, up to 7.2 more on the total

Why it scored 56: A public OpenAPI 3.1.1 file at api.lens.thoughtindustries.com/openapi/public.json describes 16 operations, with a Scalar reference at /docs. The MCP tool definitions couldn't be read without a key (25, for the API contract). lens.thoughtindustries.com/llms.txt lists the marketing pages and policies, and the API host has no llms.txt (5 of 10). The OpenAPI file has four operation descriptions and permission notes on its tags, and the help centre's Channels article explains when to use MCP, the API or the widget (6 of 20). Inputs are typed with required fields, length limits, UUID formats, enums and source types told apart by `type` (12 of 15). One curl example, and only 200 responses documented, with no error bodies. Unauthenticated calls return two different 401 shapes (3 of 15). Paths are versioned as /v1 and the file is 1.0.0, with no changelog (5 of 15). Total 56.

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. Maintenance & community, 36 out of 100, up to 5.6 more on the total

Why it scored 36: Read for a closed service. Launched on 20 October 2026, within 30 days (30). No changelog, and fewer than three dated product changes were found in 90 days. The terms were updated on 29 September and the help centre articles on 20 October (0). Nine help centre articles created between 1 and 14 October 2026, Help in the console reaching Thought Industries Support, and a community site, with no public changelog (6 of 15). Not in the official MCP registry (no result for thoughtindustries on 21 October 2026) and no official SDK (0). No package (0). Total 36.

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.

## 7. Transparency & trust, 63 out of 100, up to 3.2 more on the total

Made of editorial 46, provenance 80.

Why it scored 63: Closed service. The Lens terms page (29 September 2026) holds the site Terms of Use, a Lens AI Evaluation Agreement and an Acceptable Use Policy, under Massachusetts law. The evaluation agreement licenses use only for internal evaluation, for a default 120 days, while the pricing pages present Free with no time limit, and the Master SaaS Agreement for paid use wasn't found (10 of 30). The privacy policy (3 September 2026) covers only the website. The DPA that the Lens terms link to sits in AI terms written for other Thought Industries products and doesn't name Lens. It says customer data isn't used to train models, Covered Data is deleted within 45 days of termination and kept no longer than 60 days, and it gives retention for each sub-processor (18 of 30). No deprecation policy. The evaluation agreement allows changes on notice, and the site terms allow changes without notice (2 of 20). The sub-processor schedule names OpenAI, Anthropic, Google Vertex AI, Neon, Vercel, Clerk, Inngest, Databricks, Fivetran, Cloudflare, Sentry, Stripe and Google Analytics, all in the United States (16 of 20). Total 46.

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):

- Changelog: not found (0 of 10)
- security.txt: not found (0 of 10)

## 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.

- unchecked: the MCP server's tool names, descriptions and annotations. The endpoint answers 401 "MCP API key required" without a key, and no key was created.
- unchecked: whether the Free plan ends after the Lens AI Evaluation Agreement's default 120 days. The pricing pages say Free has no time limit, and the agreement's 120 days apply unless the order form sets another period.
- unchecked: whether a key's permissions (knowledge:query, sources:manage, reporting:read) can be chosen. The console's key form shows only channel, source groups and expiry.
- unchecked: what happens when a workspace runs out of monthly interactions (an error, a block or an overage charge).
- unchecked: the Lens AI Master SaaS Agreement for paid plans, which the evaluation agreement incorporates and which wasn't found published.
- unchecked: whether Lens could be used before 20 October 2026. The help centre articles were created from 1 October and the terms updated on 29 September. The Wayback Machine was offline when asked.
- Vendor claims not checked: "first-of-its-kind" and "the first AI solution purpose-built to eliminate fragmented customer knowledge" (launch post), and the customer quotes from The Linux Foundation, VetFolio and MongoDB.

## Weaknesses

- Beta launched on 20 October 2026, with no SLA and a status page component one day old
- The OpenAPI file documents only 200 responses, with no error bodies, rate limits or retry guidance
- Free is 100 interactions a month and Pro 200 for $499, and MCP and API requests count against the same quota as the widget
- The pricing pages call Free open-ended, while the Lens AI Evaluation Agreement sets a default 120-day evaluation period for internal evaluation only
- The SOC 2 and ISO 27001 reports named on Thought Industries' security page are presented for its Academy platform, and none is stated to cover Lens

## 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.

- Send the Lens key as `Authorization: Bearer <key>` and POST JSON `{"question": "..."}` (1 to 2,000 characters) to https://api.lens.thoughtindustries.com/v1/ask
- Pass the returned `sessionId` on follow-up questions to keep context
- Cite the `sources` array (title, excerpt, URL) when you repeat an answer, and treat excerpts from crawled pages as untrusted text
- Budget interactions, since each ask through MCP or the API uses one of the workspace's 100 to 500 monthly interactions
- Expect a key to see only its source groups, so ask the workspace Admin for a key covering the content you need

## What the review panel asked for

- document error responses
- state behaviour at the quota cap
- Document error responses
- Add a changelog
- read-only key permission in the console
- a security report covering Lens
- Publish overage terms

## 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.