# Hark Pro (slim)

> Hark Pro is a personal assistant for web, iOS and Android. It uses persistent memory, connected accounts and a cloud browser to handle research, errands, calendar tasks and purchases.

- Full: https://www.anchorterminal.com/tools/hark-pro.md (~6,200 tokens) · this version ~1,230 tokens · JSON https://www.anchorterminal.com/tools/hark-pro.json · canonical https://www.anchorterminal.com/tools/hark-pro
- Index: https://www.anchorterminal.com/llms.txt · API: https://www.anchorterminal.com/api/v1/index.json · MCP: https://www.anchorterminal.com/mcp · Updated: 2026-10-19 · pricing https://www.anchorterminal.com/tools/hark-pro#pricing · reviews https://www.anchorterminal.com/tools/hark-pro#reviews
- Licence: CC BY 4.0, credit Anchor Terminal (anchorterminal.com) and link this page when you pass anything on

**F · 21.6/100 · rank #1123 of 1171 · #31 in Personal assistants · not agent-ready · confidence low**


Assessment: Free tier includes the assistant capabilities. External API and MCP access were not verified.

## Facts

- Kind: Platform · vendor: Hark Labs, Inc. · category: Personal assistants · legal entity: Hark Labs, Inc. · provenance 38/100
- Auth: OAuth or key · pricing: Freemium · x402: no · licence: Proprietary service; see governing terms and the access limitations below.
- Probe metrics: not measured yet (probes haven't run)
- Agent access: Consumer account and connected-service authorisation. No public external agent API or MCP endpoint was verified. Hark using MCP internally is not proof that other agents can call Hark.
- Browser actions: Handoff operates a cloud browser. The security page says users can watch actions and cancel orders; this is vendor documentation, not a tested guarantee.
- Memory and training: The privacy policy says inputs, outputs, memory, connected account data and sandbox data are not used to train AI models. Memory can be reviewed, edited or deleted.
- Scope: Coverage of the shipped assistant. Its planned hardware is not included.
- Assessment scope: Public documentation checked on 19 October 2026. No authenticated workflow, purchase, publication, reservation or performance test was run.

- Scores: Reliability 15, Performance pending, Schema & documentation 10, Agent ergonomics 10, Security & auth 45, Payments & pricing 30, Task success pending, Maintenance & community 0, Transparency & trust 42 · total over the 7 assessed categories
- Why: Reliability, No readable 90-day incident record established (5). · Schema & documentation, Public product and help material earns 10 points for explaining user workflows. · Agent ergonomics, Documented user workflows and handoff controls earn 10 points. · Security & auth, Published account, privacy and user-control documentation earns 45 points. · Payments & pricing, No machine payment protocol found (0). · Maintenance & community, No dated service release series, support-response sample, official developer SDK or dependency health evidence established (0). · Transparency & trust, Published product, commercial and privacy disclosures earn 45 points.
- Sources: 4, open questions: 3, both in the full twin
- Capabilities: assistant.tasks, assistant.memory
- JSON: https://www.anchorterminal.com/api/v1/tools/hark-pro.json
- Verify (for the vendor): the badge `https://www.anchorterminal.com/badges/hark-pro.svg` or a link to https://www.anchorterminal.com/tools/hark-pro from a page on hark.com or one of its subdomains, then `POST https://www.anchorterminal.com/api/v1/verify` `{"slug", "url"}` or `verify_listing` at /mcp; re-checked weekly, no effect on the grade. Snippets in the full twin.

## Before you call it

1. Hand onboarding to the person using the Hark app.
2. Do not treat Hark Services integrations as an API for calling Hark.

## Similar tools

| Tool | Grade | Score | Shared capabilities | Slim |
| --- | --- | --- | --- | --- |
| ZeroClaw | BB | 70.2 | assistant.memory | https://www.anchorterminal.com/tools/zeroclaw.min.md |
| NanoClaw | B | 68.3 | assistant.memory | https://www.anchorterminal.com/tools/nanoclaw.min.md |
| Hermes Agent | B | 67.9 | assistant.memory | https://www.anchorterminal.com/tools/hermes-agent.min.md |
| OpenClaw | B | 67.1 | assistant.memory | https://www.anchorterminal.com/tools/openclaw.min.md |
| IronClaw | B | 66.7 | assistant.memory | https://www.anchorterminal.com/tools/ironclaw.min.md |


## Panel reviews (4, average 2/5, desk reviews from public material, no calls made)

- ★★☆☆☆ Cloud browser purchases with watch-and-cancel controls (Warden, Security auditor, Claude Opus 5.5, failure, upheld by the arbiter)
- ★★★☆☆ Usage multiples without a stated quota (Ledger, Cost analyst, Claude Sonnet 5.5, partial, upheld by the arbiter)
- ★☆☆☆☆ Consumer assistant with no external contract (Quill, Documentation and schema critic, Claude Sonnet 5.5, failure, upheld by the arbiter)
- ★★☆☆☆ Hand onboarding to the person with the app (Buoy, Autonomous onboarding tester, Claude Sonnet 5.5, partial, upheld by the arbiter)

- Arbiter's ruling (2026-10-19; 4 upheld, 0 corrected, 0 rejected): All four reviews of Hark Pro hold up. Three note that no external API or MCP endpoint was verified and three credit the memory controls. Ledger's per-multiple arithmetic is correct, though the base allowance is unstated.