How it Works For Teams For Agencies Pricing Download
Contents
  1. What Relevare does
  2. Permissions requested
  3. Network calls
  4. Independent verification
  5. Data model
  6. Org dashboard data flow
  7. Compliance posture
  8. Binary integrity
  9. Managed deployment
  10. Incident response & contact

1. What Relevare Does

Relevare is a desktop agent that observes screen and application activity on the local machine. Capture, storage, metrics, and filtering run locally. When analysis is requested, filtered text is sent to the configured AI provider. The product then surfaces workflow bottlenecks and recommendations to the user.

The product runs as a native desktop application on macOS and Windows.

Execution
User space only
No kernel extensions, no system-level drivers. Standard application-level APIs only.
Data storage
On-device only
All captured data stays in a local directory. SQLite database, local filesystem. No cloud component for captured data.
Capture scope
Foreground window only
Text extraction follows the focused window. Screen images can include the active display, including other windows that are visible on it.
Vendor access
No raw-data store
Relevare does not operate a server-side store for screenshots, recordings, or the local capture database. Filtered text is processed by the configured AI provider.

2. Permissions Requested

macOS

Permission Why it is needed What it accesses
Screen Recording Captures the content of the focused (frontmost) window to identify which applications and workflows are in use The active display. Other visible windows can appear in the screen image.
Accessibility API Extracts text from application windows without OCR where supported, faster and more accurate than pixel-based capture Text content of the active window via the macOS accessibility tree. No interaction with application logic or data outside the visible window.
Local Filesystem Reads and writes to a local SQLite database and log files Writes product data under ~/.relevare/ and uses normal app configuration and operating-system credential storage. User-approved exports can write to a chosen destination.

Windows

Permission Why it is needed
Screen capture Same as macOS: foreground window content only
UI Automation API Windows equivalent of the macOS accessibility tree for text extraction
Local filesystem Equivalent local directory for database, logs, and configuration

What is never requested

3. Network Calls

Relevare makes the minimum possible number of outbound calls. The full inventory is documented here.

Standard setup (free, ChatGPT sign-in)

Call When Payload Destination
Activation ping Once, on first launch (fire-and-forget; failures do not block startup) Signed install identity (issued at download or created on first launch), app_version, and platform. No screen data. No captured content. Relevare activation function (hosted on Supabase Edge Functions)
ChatGPT sign-in and token refresh At sign-in, and periodically to refresh the token OAuth credential exchange for the user's own ChatGPT account. No screen data. auth.openai.com
Update check Periodic background check Current version string and platform. No user data. relevare.tech (minisign-verified)
Product telemetry Only after the user opts in; org deployments may apply an admin policy Anonymous, content-free product usage events: feature interactions, activation funnel steps, intelligence pipeline cost and latency by stage, live notification fire/dismiss/ack rates, hashed device ID, plan tier. Never screen content, OCR text, prompts, LLM outputs, URLs, file paths, raw app titles, user/host names, or any captured user content. App names bucketed by category or restricted to an allowlist of well-known apps. Relevare telemetry endpoint
Crash reports Only when telemetry is enabled Sanitized stack trace with $HOME, usernames, and path components stripped. App version, OS version. No user content. Relevare crash reporting endpoint
Intelligence analysis Each analysis cycle Filtered text describing the work: app names, window titles, browser URLs with sensitive query details removed, short passages of on-screen text, selected clipboard text, and observation descriptions. Screenshots and the local database are not sent. Requests go directly from the device to OpenAI under the user's ChatGPT account; Relevare's servers do not see them. OpenAI processes the request under the user's account and OpenAI's terms. OpenAI (chatgpt.com / api.openai.com), under the user's ChatGPT sign-in
Standard setup network summary

In the standard setup, filtered analysis text goes directly from the device to OpenAI under the ChatGPT account the user signs in with, while screenshots and the local capture database stay on the device. Relevare handles identity, updates, optional finding emails, support requests, and opt-in telemetry. Relevare does not store screenshots or the local captured history on its own infrastructure. On managed org deployments, analysis can instead route through Relevare's managed AI gateway (relevare.techapi.fireworks.ai); that routing is documented for the deployment review.

Deep Research (paid add-on)

The Deep Research add-on runs live web search during analysis so recommendations reflect the current tool landscape rather than stale training data. It adds the two calls below on top of the standard inventory.

Call When Payload Destination
Web search queries During each analysis cycle when deep research is enabled Search queries derived from detected bottleneck context. No screen content, no employee names, no file contents. Relevare research proxy
Research synthesis During each analysis cycle Same query context as above Relevare LLM endpoint

Deep research does not send screenshots or the local database. Search queries are filtered and shortened, but they can still reflect workflow context and should be included in a privacy review.

Organization tier

Organization enrollment adds signed policy and entitlement checks. Any dashboard or aggregate-data integration is documented and reviewed for that deployment; the public desktop release does not enable captured-data cloud sync.

Call When Payload Destination
Enrollment and policy refresh At enrollment and when signed policy is refreshed Install identity, organization code or signed configuration, and product version. No screenshots or local database. Relevare identity service
Aggregate insight delivery Only when enabled for a contracted organization deployment Aggregate findings without individual attribution. The exact schema, retention, and subprocessors are supplied for review before deployment. Relevare organization service

Employee visibility and non-attribution are product requirements: employees must be able to see what leadership sees, and leadership does not receive named individual activity.

4. Independent Verification

You do not need to take our word for the network call inventory. You can verify it yourself.

Packet capture (macOS)

# Install Wireshark or use the built-in tcpdump
# Run during a full Relevare session including first launch

sudo tcpdump -i any -w relevare_capture.pcap host [your-machine-ip]

# Or filter to Relevare's process specifically:
lsof -i -n -P | grep -i relevare

Open the capture in Wireshark and filter by the Relevare process. A standard installation contacts Relevare for identity and updates, and OpenAI (auth.openai.com, chatgpt.com, api.openai.com) for ChatGPT sign-in and AI analysis under the user's own account. Telemetry and crash reporting appear only after opt-in. Deep Research adds web-search and research-synthesis calls when enabled.

On-device data inspection

# Database location
~/.relevare/db.sqlite

# The database is SQLCipher-encrypted. Inspect through Relevare's local
# APIs or with a SQLCipher-capable tool and the local .db-key.

# Log files
~/.relevare/relevare.YYYY-MM-DD.0.log

# Watch what the pipeline is doing in real time
tail -f ~/.relevare/relevare.*.log

You can verify directly what is stored, what the intelligence pipeline produces, and confirm that no raw screen content persists beyond the retention window.

5. Data Model

What is captured

What is not captured

No microphone or camera capture. Relevare does not store a raw stream of individual key codes or continuous mouse movement. Screen images can still contain any window visible on the active display.

What is stored locally (tiered retention)

The retention scheduler runs hourly with per-layer TTLs that reflect how the intelligence pipeline actually consumes each data type:

Data layer Default TTL Why this TTL
JPEG snapshots 1 day Largest disk consumer. Never read by the intelligence pipeline.
Accessibility tree JSON (raw) 1 day Removed after text extraction.
Frame text (accessibility text + OCR) Analysis-aware Kept at least 24 hours; normally removed after completed analysis, with a 3-day fallback and 14-day hard ceiling.
Frame metadata (timestamps, app/window, hashes) 7 days Supports recent metrics and profile inference.
Intelligence metadata (insights, ledger, summaries) Forever Product output; contains no raw captured content.

The scheduler removes files, clears raw columns and side tables, then deletes expired metadata rows. Intelligence tables are not purged automatically.

Encryption at rest

The SQLite database is encrypted at rest using SQLCipher (AES-256). The key is auto-generated on first launch (32 random bytes), stored at ~/.relevare/.db-key with mode 0600 on Unix, and applied via PRAGMA key. Without it, sqlite3 ~/.relevare/db.sqlite returns file is not a database. Existing unencrypted databases from earlier releases are migrated transparently on upgrade.

Exclusions and limits

Retention configuration

Parameter Default Notes
--retention-days 0 Tiered defaults above 0 disables every layer (keep forever). Any non-zero value uses the tiered defaults.
Intelligence metadata Kept indefinitely Never purged automatically
Product analytics & error telemetry Off by default; users may opt in and org deployments may apply an admin policy. Anonymous, content-free product usage events and sanitized crash reports. Never contains screen content, OCR text, prompts, LLM outputs, URLs, file paths, raw app titles, or user/host names. Inspectable via in-app telemetry viewer before send. See Section 3 (Network Calls) for full payload detail.

IT administrators can configure retention at deployment time via CLI flag or environment variable. Custom uniform TTLs are also supported by setting individual layer values in the deployment config.

Local API authentication

The HTTP API at 127.0.0.1:3030 requires authentication for all mutation endpoints (POST, PUT, PATCH, DELETE). A 32-byte API key is auto-generated on first launch and stored at ~/.relevare/api-key.txt. Read endpoints (GET) and the /health liveness check are exempt. This prevents cross-process abuse from other local applications, including malicious browser extensions and CSRF from a logged-in tab.

Credential storage

ChatGPT credentials are stored in the operating system's credential manager when supported. Local identity and license files under ~/.relevare/ are restricted to the current operating-system account. Build-time managed AI signing secrets are not embedded in public release binaries.

Solution build handoff

Relevare can prepare a build brief and hand it to a coding agent chosen by the user. The public build flow does not expose a remote tool that silently installs or enables local automations. The coding agent works in its own environment and uses its normal approval controls.

6. Org Dashboard Data Flow

The org dashboard is an optional paid feature. Understanding how it works is important for evaluating the privacy implications for employees.

What rolls up

Only anonymized, aggregated intelligence patterns are transmitted to the org dashboard:

What never rolls up

Employee visibility guarantee

Employees can see exactly what their manager sees in the org dashboard. There is no hidden view available to administrators that is not also visible to the employee. This is a product constraint, not a setting.

7. Compliance Posture

Current status as of August 2, 2026:

Framework Status Notes
Privacy policy published Plain-language plus formal GDPR/CCPA sections at /privacy.
Terms of service published Available at /terms.
SOC 2 Type II not certified Planned; no certification is claimed.
HIPAA not certified Relevare does not currently offer a BAA.
ISO 27001 planned Planned post-SOC 2
GDPR program in progress On-device architecture reduces exposure; legal and operational obligations still apply.
CCPA minimized exposure On-device architecture minimizes exposure. Privacy policy documents consumer rights.
Honest compliance posture

We will not claim compliance we have not achieved. If your organization requires a specific certification before deployment, contact us to discuss timeline and interim controls.

8. Binary Integrity

All Relevare release binaries are:

Do not deploy binaries obtained from any other source. If your MDM or software distribution system requires a verified hash, contact us at [email protected] and we will provide SHA-256 checksums for each release.

9. Managed Deployment

macOS (Jamf / other MDM)

Relevare can be deployed silently via MDM. The signed and notarized .pkg installer accepts the following deployment configuration:

Windows (Intune / SCCM)

MSI installer with equivalent silent deployment parameters. Contact [email protected] for deployment documentation.

Employee consent

Silent deployment without employee knowledge is never acceptable. All deployment modes include a mandatory employee consent flow. Employees must be informed that Relevare is running and must have visibility into what data is being collected.

10. Incident Response & Contact

Contact Purpose
[email protected] Security vulnerabilities, data concerns, compliance questions
[email protected] IT deployment support, MDM configuration, managed deployment

For security vulnerability disclosure, we follow a 90-day responsible disclosure policy. Please contact [email protected] before public disclosure. We acknowledge reports within 48 hours and triage within 7 days.