Lindy

# Fix list: Lindy

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/lindy, the October 2026 research run, assessed 12 October 2026. Grade F, 32.3 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 Lindy: 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, 24 out of 100, up to 15.2 more on the total

Why it scored 24: Graded as a hosted service, on the assistant itself, since no public API was found. No status page is linked from the site, the docs or the security page (0). With no page there is no incident history to read (5). No request limits are published. The pricing page gives credit ranges for a task (2 to 250, 250 to 1,000 and 1,000 to 2,500) and the memory page caps an upload at 100 files (4 of 15). The docs describe what happens at a limit, a pause with a message when credits run out and a prompt to continue or stop when a task draws more than usual, with no retry guidance for failed writes (5 of 15). No SLA was found on the pricing page, the terms or the docs (0). The product is not marked beta, and Microsoft Teams is marked coming soon (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. Schema & documentation, 32 out of 100, up to 11.1 more on the total

Why it scored 32: There is no API, so this is read on the documentation of how the assistant behaves. No machine-readable contract (0). The docs publish `llms.txt` and a Markdown copy of every page (10). The pages on integrations, credentials, MCP servers, memory and routines say who can see and change each thing and where approval applies, in tables (11 of 20). Skills have a documented file format (`SKILL.md` in a `.skills/` folder), and nothing else is typed (2 of 15). The docs give example requests and troubleshooting steps, with no list of errors (6 of 15). The public changelog stops at 21 October 2025, and the docs and the pricing page disagree on plan names and on the number of integrations (3 of 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.

## 3. Agent ergonomics, 43 out of 100, up to 9.3 more on the total

Why it scored 43: Read on how a person directs the assistant. Skills load their instructions on demand, and MCP tools can be switched off one by one or all at once, though every tool arrives on (12 of 25). Event triggers take a plain-language filter and a routine has a named destination. No control over the length of a reply was found (6 of 20). Connections show whether they are healthy or need reconnecting, the MCP page explains a zero tool count, and a usage warning asks before a costly task continues (8 of 20). A dry run fires a routine on real data and suppresses every send, post and write, and memory files keep every version. No guidance on retries was found (10 of 20). Routines default to daily at 9:00, setup is by sign-in with no keys, and there is no SDK or API (7 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.

## 4. Security & auth, 49 out of 100, up to 8.9 more on the total

Why it scored 49: Accounts connect by OAuth or an API key, personal by default and shared only by an owner or admin. Saved keys are write-only and can be rotated and revoked. Lindy has no scope choice, since every action of a tool turns on at once, and a shared connection lets everyone's Lindy act as that account (20 of 30). No deduction was taken for the MCP dialogue's option to send a server's static credential as a query parameter, because the secret is the third-party server's and a header is also available. Each integration can require approval or bar writes, but the docs say this applies in shared Slack threads only, that web chat, Slack direct messages, iMessage and SMS are not guarded, and that saved credentials are not covered. The default setting is not stated (9 of 20). Lindy reads mail, Slack, web pages and MCP results, and no prompt-injection guidance was found. Personal context is kept out of shared channels per the docs (2 of 15). The security page says every action is logged, and audit logs are an Enterprise item (9 of 15). The site claims SOC 2 Type II audited by Johanson Group, penetration tests and an incident response plan. The trust centre answered with a bot check and was not read. No security.txt, disclosure policy, bug bounty or advisory page was found (9 of 20).

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. Payments & pricing, 35 out of 100, up to 8.1 more on the total

Why it scored 35: No x402, MPP or L402 (0). The pricing page is public and gives a seat price, fifteen monthly credit amounts and a top-up price of $10 per 1,000 credits. A task's cost is given only as a range, and the docs name Plus, Pro and Max plans priced inside the app (15 of 20). New accounts get $50 of credit for seven days with no card (20). A person signs in with Google or Microsoft in a browser (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.

## 6. Maintenance & community, 9 out of 100, up to 8 more on the total

Why it scored 9: The public changelog's newest entry is 21 October 2025, 356 days before this check (0). No dated entry in the last 90 days (0). The docs repository on GitHub had 115 commits between 30 August and 9 October 2026, and the terms and privacy policy were revised on 24 and 28 September 2026, so the product is being worked on without release notes. Support is by email and phone, with group onboarding sessions. No public issue tracker (9 of 25). No SDK and no MCP registry entry, which a client would not have (0). No package to assess. The Chrome Web Store page redirected to a consent page and was not read (0).

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, 52 out of 100, up to 4.2 more on the total

Made of editorial 36, provenance 68.

Why it scored 52: Closed service. The terms are public and name the company, but read as a general website template. They forbid automated use and use for any commercial enterprise in a product sold to companies for automating work (10 of 30). The privacy policy and the security page agree that data is not sold and not used to train models, and that Google data is not used for AI training. Retention is as long as necessary with no period, health data is listed as a category processed, and no DPA was linked (14 of 30). No deprecation policy or dated notice was found (0). Processors are named by category in the privacy policy, with no locations (12 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: lindy.ai, registered 2019-05-16 (7 years) (11 of 15)
- Terms of service: read, states 6 of the 7 things a reader expects, and has 3 clauses that cost points (3.1 of 10)
- Privacy policy: read, states 7 of the 8 things a reader expects (9.3 of 10)
- Status page: not found (0 of 10)
- security.txt: not found (0 of 10)

## Deductions

Each comes off the total. A fixed and documented problem counts for less at the next check.

- 2026-10-12. The pricing page and the docs pricing page answer the question of whether Lindy does anything without approval with No, and say anything with outside impact waits for approval. The integrations page says approval settings apply in shared Slack threads only, that web chat, Slack direct messages, iMessage and SMS are not guarded, and that an integration can be set to always allow writes (https://www.lindy.ai/pricing, https://docs.lindy.ai/pricing.md, https://docs.lindy.ai/integrations/overview.md). 3 points for a misleading claim.

## 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.

- Kind is `platform`, the briefs' kind for a web or phone app, because Lindy is a hosted product with no public API, MCP server or SDK. Hosted lines were used for Reliability and the hosted rubric for Payments.
- Lindy is a team product with a personal assistant half. The home page calls it an AI teammate for the whole team, and the enterprise page and docs describe a private personal Lindy for one user's mail, calendar and meetings. It was graded in personal-assistants on that half.
- The lead's phrase about an AI employee that connects to all your tools was not found on the home page as read today, which says AI teammate. The pricing FAQ says AI employee.
- The terms forbid any automated use of the system, robots and scrapers. Recorded as a fact with no deduction. It matters before any probe is run.
- Maintenance scores recency on the public changelog, which stops at 21 October 2025. Scored on the docs repository's last commit of 9 October 2026 the recency line would be 30.
- Whether the default approval setting for a new integration is always allow or require approval is not stated in the pages read.
- Where OAuth tokens are stored, and whether Pipedream holds them, is not stated. The privacy policy lists Pipedream and WorkOS under integrations and authentication.
- The Chrome extension page shows a model picker with Gemini and DeepSeek models in an illustration. The privacy policy's list of AI providers has neither Google as a model provider nor DeepSeek.
- The docs say a saved credential cannot be read back through the API. No API documentation was found.
- unchecked: the trust centre at app.drata.com, which answered robots.txt with a bot check (403). The SOC 2 report, sub-processor list and any DPA there are unread.
- unchecked: the Chrome Web Store page, which redirected to a Google consent page, so the extension's version, date and permissions are unread.
- unchecked: plan prices for Plus, Pro and Max, which the docs say are shown only in the app's billing settings.
- unchecked: the Slack app's permission scopes, which appear only on Slack's approval screen.
- unchecked: the integrations catalogue, the blog and the case studies, which were not read.

## Weaknesses

- The integrations page says approval settings apply in shared Slack threads only. Web chat, Slack direct messages, iMessage and SMS are not covered, and reads are never covered.
- The pricing page and the docs pricing page both answer that Lindy does nothing with outside impact without approval, which the integrations page contradicts.
- No status page, uptime record or SLA was found on the site or in the docs.
- The public changelog's newest entry is dated 21 October 2025, and it describes an agent builder the current docs no longer cover.
- Deleting a memory file keeps its earlier versions, and no way to purge them or a retention period was found. The privacy policy gives no period either.
- The terms of 24 September 2026 forbid automated use of the system and use for any commercial enterprise, in a product sold to companies. This matters before any probe is run.

## 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.

- Treat a request sent by direct message, text or web chat as unguarded. Approval prompts appear in shared Slack threads only.
- Run a routine as a dry run before turning it on. A dry run reads real data and sends, posts and writes nothing.
- Turn off unused tools after adding an MCP server. Every supported tool arrives switched on, and a large server adds more than a hundred.
- Expect a billed seat for anyone who mentions Lindy in Slack. A teammate who joins that way has seven free days first.
- Connect only hosted MCP servers on public HTTPS with streamable HTTP, the one kind the docs describe.

## 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.