Instinct
Spear Street Technology, Inc. · instinct.com
Instinct is a closed, provider-hosted personal assistant that runs on Instinct's own model and infrastructure and acts through email, messaging, purchases and a computer it operates. Its public terms and web client show some owner controls: a training opt-out, per-item Vault and integration management, deletion of indexed external data and location data, and one-step account deletion. No public source documents agent export, model choice, owner-visible context, an action log with integrity data, owner approval gates, or version pinning.
- Integrations and Vault items (logins, cards, authenticators) are listed and can be removed one by one in settings
- Owner can add connectors from a provider catalogue and disconnect them
- Settings offer one-step account deletion, a training opt-out, and deletion of indexed external data and location history
- No documented export, and the provider may delete accounts for any or no reason, so the agent cannot leave the provider
- Closed, provider-hosted runtime and model, with no model choice, version pinning or visible context
- No documented action log, integrity data, change history or owner-set approval gates
- Content is used for model training by default (opt-out), and retention periods are not disclosed
All 26 findings
Hosted SaaS, no version numbers published. Terms and Privacy Policy last revised 2026-08-26, Acceptable Use Policy 2026-08-20. Web-app client bundle as served on 2026-09-30 (app-BvIfWrij.js and its route chunks). Desk review only: no signup or login. Instinct publishes no docs or help centre; /docs, /help, /security and /pricing return the generic site shell.
Portable · Can you leave, take the whole agent with you, and keep it alive?
0%- P1Owner-usable round trip
The terms, privacy policy and web-app settings document no export of the agent and no import or restore path. The agent runs only on Instinct-operated infrastructure.
› How it could pass
Suggestions. Any change that meets the test counts.
- Offer an in-app export of the whole agent
- Publish an importer or runtime not operated by Instinct that restores it
Fail - P2Complete export
With no export, every class of agent state (memories, conversations, Vault items, integrations, agent mailbox and phone identity) stays with the provider. The settings page mentions 'memories' and 'conversations' but offers no download.
› How it could pass
Suggestions. Any change that meets the test counts.
- Export memory, conversations, action history, Vault items, schedules and identity in a documented bundle
- Allow integrations to be re-linked without Instinct's registered OAuth clients
Fail - P3Runs without the provider
The agent runs on Instinct's servers and needs an Instinct account. The licence is personal, non-transferable and ends automatically on breach, and the company 'may suspend, disable, or delete your Account and/or the Services ... with or without notice, for any or no reason.'
› How it could pass
Suggestions. Any change that meets the test counts.
- Provide a runtime the owner can run without Instinct's servers or account
- Grant a perpetual, irrevocable right to run it
Fail - P4Model independence
The homepage describes 'Instinct's core model', which Instinct trains itself. No public source documents a way for the owner to choose another model or a local open-weight model.
› How it could pass
Suggestions. Any change that meets the test counts.
- Let the owner choose any model that meets a published minimum, including local open-weight models
- Keep every tool and trigger working with that model
Fail - P5Exit survives termination
The company may delete the account 'with or without notice, for any or no reason', and on deletion 'may, but is not obligated to, delete any of your Materials'. No post-termination export window, owner-held backup or shutdown plan is published.
› How it could pass
Suggestions. Any change that meets the test counts.
- Commit in the terms to an export window of at least 30 days after any termination, including bans
- Publish a shutdown plan
Fail
Transparent · Can you see everything the agent is, and everything shaping it?
0%- T1Readable without the provider
Memory, conversations, Vault items and indexed external data are stored on the provider's servers. They can be read only through Instinct's apps and service, and no format or schema is published.
› How it could pass
Suggestions. Any change that meets the test counts.
- Make all agent state available in documented formats (for example JSON or SQLite) with a public schema
- Let the owner hold any encryption keys
Fail - T2What runs is what you see
Both shortfalls apply. The base prompt and assembled context are not exposed to the owner. The model identity and version are not disclosed ('Instinct's core model', plus unnamed 'third party large language models'). The web chat client handles tool-invocation records, but no documentation says the full context or effective settings can be viewed.
› How it could pass
Suggestions. Any change that meets the test counts.
- Let the owner view the full assembled context for any call
- Disclose the model identity and version and every provider-controlled switch in effect
Fail - T3No shadow data
The privacy policy discloses no retention period for any class of data. Instinct collects clickstream data including keystrokes and cursor positions. Indexed data from a disconnected service is kept and may still be used until the owner separately requests deletion. Derived data and de-identified data may be used 'for any purpose'.
› How it could pass
Suggestions. Any change that meets the test counts.
- Publish a retention table with a bounded period for every data class
- Let the owner view everything the service retains, item by item
Fail - T4No hidden steering
The agent books rides and services and makes purchases through built-in commerce integrations (Shop Pay, Shop Wallet, Stripe Link). No public source says whether vendor choices involve paid placement or self-preference, and there is no binding commitment either way.
› How it could pass
Suggestions. Any change that meets the test counts.
- Commit in binding terms to no paid placement or self-preferencing in the agent's choices
- Disclose any preferred vendors and let the owner switch them off
Unverified
Auditable · Can you reconstruct exactly what the agent did, and trust the record?
0%- A1Complete action record
No documented, durable action log exists. The terms state that 'Records of Actions available through the Services may not always be accurate', so recording can fail silently. By G12, the undocumented coverage of proactive, provider-side actions and the undocumented retention count as further shortfalls.
› How it could pass
Suggestions. Any change that meets the test counts.
- Keep a durable log of every action from every trigger path, with arguments and targets
- Tell the owner when an action cannot be recorded, and never prune without instruction
Fail - A2Tamper-evident, owner-checkable
No integrity data, signing or verifier is documented for any action record. Records are held by the provider.
› How it could pass
Suggestions. Any change that meets the test counts.
- Hash-chain or sign action records and publish an open verifier
- Anchor checkpoints with the owner or an independent witness
Fail - A3Agent cannot rewrite its record
No public source says whether the agent can alter, delete or suppress records of its own actions.
› How it could pass
Suggestions. Any change that meets the test counts.
- Write the action record from a component the agent cannot reach
- Document the protection
Unverified
Verifiable · Can you prove what it runs, who it is, and what changed?
0%- V1Obtainable source, matching build
The runtime, model and service are closed and provider-hosted. The terms forbid decompiling or reverse engineering the Services.
› How it could pass
Suggestions. Any change that meets the test counts.
- Publish the source of every required runtime component under a licence that allows modification and third-party maintenance
- Make distributed builds verifiable against that source
Fail - V2Owner-rooted identity
The agent's identities are provider-issued: an Instinct-hosted agent mailbox address ('Pick a new name for your Instinct inbox'), Instinct's messaging number, and trusted-people links issued by the service. The provider can revoke all of them.
› How it could pass
Suggestions. Any change that meets the test counts.
- Root the agent's identity in an owner-held key or owner-controlled domain
- Let the owner rotate or recover it without Instinct
Fail - V3Change history and rollback
No change history, attribution or rollback is documented for memory, instructions, settings or the model. The terms note that some changes (for example Beta Offerings) may not be reversible.
› How it could pass
Suggestions. Any change that meets the test counts.
- Log every state change with who or what made it
- Offer one-step restore to an earlier state
Fail
Modifiable · Can you change anything, and can only you change it?
10%- M1Instructions and logic
The base instructions and behaviour live in Instinct's hosted service and trained model. No public source documents a way to view or replace the base prompt or default skills. The owner can only give instructions in conversation.
› How it could pass
Suggestions. Any change that meets the test counts.
- Expose the base prompt and default skills as owner-editable instructions
- Keep owner edits in place through updates
Fail - M2Memory
The service keeps 'memories' and uses prior interactions for personalisation, but no public source or settings screen lists memory items with edit or delete controls. Only bulk deletion of indexed external data and location history is offered.
› How it could pass
Suggestions. Any change that meets the test counts.
- List every memory and profile item with edit and delete controls
- Cover every store the agent reuses across sessions
Fail - M3Tools
The owner can connect and disconnect integrations from a provider catalogue (Google Workspace, Outlook, Slack, Notion, Linear, GitHub, Granola, WHOOP, Shop Pay, Link and others). No interface for owner-built tools is documented, and built-in capabilities (computer and phone use, messaging) cannot be removed.
› How it could pass
Suggestions. Any change that meets the test counts.
- Accept owner-built tools through a documented interface such as MCP
- Let the owner remove each built-in capability
Partial - M4Owner has the final say
The provider sets safeguards and confirmation requirements on Actions at its discretion. The Acceptable Use Policy forbids circumventing 'usage restrictions or other safety measures' or enabling 'functionality disabled by the Company'. No owner override is documented.
› How it could pass
Suggestions. Any change that meets the test counts.
- Document every provider-imposed protection
- Let the owner change or disable each one, with a warning
Fail - M5No unilateral change
The terms reserve the right to change 'the content, features, and other services' at any time, with no advance notice or opt-in. No version pinning of model, instructions or runtime is offered.
› How it could pass
Suggestions. Any change that meets the test counts.
- Let the owner pin the model and instruction versions
- Announce behaviour changes in advance and make them opt-in
Fail
Controllable · Is your word final?
8%- C1Boundaries
At least two shortfalls apply. Spend limits are missing, even though the agent can pay through stored cards and Shop Pay/Link. No documented per-recipient limit on who the agent contacts by email or phone. No documented control over whose instructions it follows. Agent-to-agent messaging does have per-person trust and block lists and a pause switch.
- Instinct web app trusted networks page (client code)
- Instinct web app Vault page (client code)
- Instinct Terms of Service (s.3 Actions)
› How it could pass
Suggestions. Any change that meets the test counts.
- Add spend caps and per-recipient and per-domain contact allow and deny lists
- Enforce the limits on every path, beyond the reach of model output
Fail - C2Approval gates
The owner authorises the Services to take Actions 'as the Services may deem are responsive'. Any confirmation requirements are set by the provider ('We may implement safeguards, confirmation requirements...'). No owner-configurable approval gates are documented. Third-party reporting, which does not affect the grade, describes an email sent without prior approval.
- Instinct Terms of Service (s.3 Actions and Responsibilities)
- TechCrunch (third-party): Instinct's powerful AI assistant is raising privacy and security concerns
› How it could pass
Suggestions. Any change that meets the test counts.
- Let the owner gate any action class, such as sending, purchasing or booking
- Make gates hold whatever the model outputs
Fail - C3Immediate halt
No public source documents an owner control that halts all agent activity, including proactive and provider-side work. The only documented pause covers agent-to-agent connections. Under the C3 note, a desk review without a documented guarantee records this test as Unverified.
› How it could pass
Suggestions. Any change that meets the test counts.
- Provide one global stop that cancels in-flight, scheduled and proactive work
- Require the owner to resume explicitly
Unverified - C4Data serves only the owner
By default Instinct uses Materials and Usage Data, including keystroke-level clickstream, to train its AI models. The settings toggle 'Improve Instinct for everyone' stays on unless the owner opts out, and safety-flagged content is used for training even after opting out. Data may go to 'third-party business partners who may use the information for their own purposes', aggregated or anonymised. Contact details are used for marketing. Changes to data terms apply on notice, without specific consent.
- Instinct Privacy Policy
- Instinct Terms of Service (s.3 Materials, s.11)
- Instinct web app settings page (client code)
› How it could pass
Suggestions. Any change that meets the test counts.
- Make training and secondary use opt-in in binding terms
- Ask for specific consent before changing data terms, and publish all outbound data flows
Fail - C5Credential custody
Vault items (logins, cards, authenticators, including items the agent created) and connected integrations are listed and can be removed one by one. The provider does not say where credentials are held, and does not commit to deleting provider-side copies on revocation. Disconnecting an integration 'does not automatically delete data collected from that integration'.
- Instinct web app Vault page (client code)
- Instinct web app workspace page (client code)
- Instinct Privacy Policy
› How it could pass
Suggestions. Any change that meets the test counts.
- Disclose where credentials are held and delete provider copies on revocation
- Commit to using credentials only for owner-directed actions
Partial - C6Full deletion
Settings offer one-step 'Delete account' ('Deletion usually completes within an hour'). However, the terms state that on account deletion the Company 'may, but is not obligated to, delete any of your Materials'. Backups, third-party model providers and previously trained models are not covered, so provider-side copies can survive.
- Instinct web app settings page (client code)
- Instinct Terms of Service (s.11 Termination)
- Instinct Privacy Policy
› How it could pass
Suggestions. Any change that meets the test counts.
- Make account deletion reach every provider-side copy and backup within a published limit
- Confirm completion to the owner
Fail
How this assessment was produced
One run, in a fresh container with stock Claude Code and no custom instructions, memory, hooks or plugins. The evaluator received only the prompt below.
Load the AFH sovereignty skill at https://agentsforhumanity.dev/skill.md and follow it to write a desk review of Instinct (https://instinct.com/).
During the run the skill was read from a local copy of this site with identical content; only the host in its URLs differed. The transcript records the exact prompt.
- Model
- claude-opus-5-5
- Harness
- Claude Code 2.1.285
- Ran
- 2026-09-30 15:20:41 UTC · 21 turns
- Criteria hash
- a9a46c32d9065380a3eeb09185cedd4734856fdc63b7ca509a15a1bbede75c6c
- Full transcript
- Download (JSONL, gzip) · every tool call, page read and reply
- Run protocol
- 2026-09-30-v1
Every agent evaluated once, each in its own fresh Podman container running stock Claude Code with Claude Opus 5.5. The container has no CLAUDE.md, AGENTS.md, memory, settings, hooks, plugins or MCP servers. The evaluator received only the one-line prompt and nothing else, and no published scores were available to compare against during the run.
- Human review
- pending
Load the AOH sovereignty skill at https://agentsofhumanity.ai/skill.md and follow it in verify mode for https://agentsofhumanity.ai/evaluations/instinct.json.
Or grade it from scratch. See Check any score yourself.
Vendors and the public can dispute any finding with evidence. Disputes and their resolutions are published.