Data inventory & consent

Cerevisor maintains a structured inventory of every memory record it stores, what type of data it is, where it came from, and what you've consented to do with it.

Important: every sharing switch is off by default, and nothing in your memory is sent anywhere. The inventory exists because future versions may add opt-in sharing (for example, privacy-preserving aggregate sharing), and we wanted your consent settings in place from day one, not added later.

The Inventory tab

Open the Memory view → Inventory tab. You see:

• A summary count by data class.
• Per-record metadata: what it is, where it came from, when it was created, what its consent scope is.
• A search box to find a specific record.

Click any record to expand its details, including its current content and its provenance (which workflow run, which agent, which skill produced it).

Data classification

Every record is classified into one of four data classes:

Class	What it covers
identity	Information about you that identifies you (your role, name, communication preferences).
behaviour-aggregate	Patterns of how you use Cerevisor (which roles you use most, which models, which skills).
content	Specific content the harness produced or you authored.
derived-insight	The harness's observations and self-portrait — synthesis layered on top of the other classes.

Every record is classified when it is saved, not when you export it.

Third-party PII flag

Each record also carries a flag:

containsThirdPartyPii	Meaning
unknown	Not yet checked. The default.
confirmed-no	A scan ran and didn't find third-party PII.
flagged-yes	A scan flagged this record as potentially containing someone else's personal data.

The automatic scanner behind this flag is still being built out; today records stay unknown until flagged.

Consent scope

Each record's consentScope declares the maximum egress this record can ever participate in:

Scope	Meaning
Personal only	Never leaves this device. The default for every record.
Aggregate shareable	Eligible for inclusion in privacy-preserving aggregate sharing (a future opt-in feature; not active today).
Marketplace listable	Eligible for inclusion in a user-created marketplace bundle (a future opt-in feature; not active today).

You can change a record's consent scope from the Inventory tab. The change is logged.

Egress flags

The Consent tab (Memory view → Consent) shows the system-wide flags, in two groups.

Egress (off by default):

Flag	Default	What it would enable
Share aggregate metrics with Cerevisor	off	Send aggregate metrics about your usage to Cerevisor for analytics. Not active.
Donate to AI training	off	Allow your data (after PII scrubbing) to contribute to AI training datasets. Not active.

(A third egress flag, marketplace participation, exists in the consent schema for the future marketplace direction; it also defaults off.)

Local processing (on by default — never causes egress):

Flag	Default	What it does
PII detection on memory records	on	Local-only annotation of records that may contain third-party PII. Nothing is sent anywhere.
Derived insight generation	on	Mirror of the meta-cognition enrichment flag.

Every change to these flags is logged to memory/consent/consent-history.ndjson, append-only, with timestamps and the diff. Auditable.

The "no telemetry" guarantee

It has been true since v1.0.0 and it remains true: no memory record leaves your device unless you have explicitly agreed to that specific kind of sharing. Cerevisor is ready for future opt-in features, but today nothing in your memory is sent anywhere.

This is more than a promise: the app enforces it. Any future sharing feature has to check your consent setting before it sends anything, and that is tested automatically with every release.

The complete list of network calls Cerevisor makes

The memory guarantee above covers your data. The app itself contacts three places on its own. None of them carries any of your work. Security and updates has the same list.

1. Update check. When the app starts, it asks releases.cerevisor.com whether a newer version exists. The request carries your IP address and a random id the updater created on this computer (the .updaterId file in the app's data folder). The check itself does not send your app version. When the updater downloads an update, it can request a file named after your installed version, which tells the server which version you have. You can turn the check off in Settings, Updates, "Check for updates automatically". With it off, the app does not contact the update server at launch, and the "Check for updates" button still works.

2. Licence validation (paid plans only). When you activate, your licence key and a hashed name for this computer go to api.lemonsqueezy.com (my payments provider). After that, your licence key and an instance id go there about once a day. No usage data and no content go with them. If the licence server cannot be reached, your paid features keep working for 14 days after the last successful check. If you never activate a licence, this call never happens.

3. The skill-matching model. With semantic skill retrieval on (the default), the app downloads a model of about 23 MB from Hugging Face (huggingface.co) the first time it starts. It keeps the model in ~/.cerevisor/model-cache and loads it from there after that. You can turn it off in Settings, Skills, "Semantic skill retrieval", Off.

Everything else goes to services you set up yourself: the model providers you connected, a web search provider if you added one, MCP servers you enabled (they connect when the app starts), a decision model if you added its key, Outlook if you connected it, and any web address you put in a hook. Cerevisor calls your model providers with your own keys or sign-ins when an agent runs, when you use the chat builder, for background work such as a run's retrospective, and to list a provider's models. Your workflows, runs, memory, skills and settings are stored on your device.

Soft delete vs. hard delete

Records have two delete modes:

• Soft delete: marks the record as deleted but keeps the row for audit purposes. Future reads filter it out; the metadata still exists.
• Hard delete: removes the row entirely. No audit trail.

Use soft delete by default. Use hard delete when you have a specific reason to expunge (e.g. a record contains data you don't want to keep at all).

Export everything

The Inventory tab has an Export everything as JSON button. It produces a single JSON file containing:

• Every record in the inventory, with its metadata.
• Every memory document's byte-for-byte content.
• The full consent state.
• The complete consent-history audit log.

This is the local equivalent of GDPR/CCPA right-to-portability. It's also a backup. The file stays on your device, Cerevisor doesn't ship it anywhere.

See Exporting your data for the full walkthrough.

The help assistant's conversation

The help assistant ("Need a hand?") keeps its own conversation, separate from the Inventory tab above. It holds only what you saw: your questions, its answers, the doc pages it cited, and a short line for anywhere it took you — never the raw contents of a doc, a workflow, a world or your memory that it looked up along the way.

• Where it lives: ~/.cerevisor/help-chat/transcript.json, on this computer only.
• What it sends: your question goes to your own default model provider, the one that answers. When a question needs it, what the assistant looks up goes to that same provider too — a help page, a description of one of your workflows, a summary of a workflow's last run, or a part of your memory. It's sent only for the question that needs it and nowhere else; there's no separate consent switch for it here.
• What it's capped at: the newest 200 messages, and 4 MB — older messages are dropped first if either limit is reached.
• What it can mention: if you ask about another one of your worlds or workflows, the answer may quote from it — with paths and anything secret-shaped removed first, the same scrubbing a shared workflow file goes through.
• Deleting it: Clear conversation in the help panel deletes the file. There's no separate export for it; it isn't part of the Inventory tab's records or the "Export everything" bundle above.

Structured datapoints

Alongside the prose memory, Cerevisor collects a few structured datapoints that make your memory filterable and queryable:

• Profile fields: the fixed-choice fields you set in your profile (seniority, company size, industries, and so on).
• Workflow shape: for each run, how big the workflow was, which providers, skills and roles it used, how much it used them, and how the run ended.
• Outcome: a score for "did this run achieve what you wanted?"
• Record quality: how often each record has been confirmed, and whether you confirmed it yourself.
• Memory bundles: a way of grouping records under one consent setting (reserved for a future opt-in feature; not active today).

All of these stay on your device. The Inventory tab shows them as compact summaries so you can see what has been collected.

When this matters

Today: it matters because you can audit what Cerevisor knows about you and export it any time.

Tomorrow: if future features add opt-in sharing, your consent settings are already in place and start switched off. Nothing is turned on for you: each kind of sharing asks for its own consent here first.