SourceLace
The SourceLace Trust Layer

Your rules apply before the AI sees anything.

SourceLace acts as each person, with every action approved and on the record. Every question, change and agent run, in the SourceLace app, by an AI agent or from your own AI app, passes through the same layers of the SourceLace Trust Layer, in the same order, every time.

For security and IT

What your security team needs to approve.

Five things, in plain terms. Each says what SourceLace does, not how.

  • Masking of sensitive data. Your AI works without seeing your sensitive data. The right people still see the real values. Your admins choose which fields are hidden or masked. Photos are the exception: they are read by the AI only if an admin allows them, and data protection rules cannot check what is in a picture.
  • Approvals before changes. Read-only by default. A change happens only where an admin allows it, and only after a person has seen it and said yes. Agents earn trust before they act alone: tests, a trial run, then approvals.
  • A tamper-evident audit trail. Every question, change and agent run is on the record: who, what and when, never the data itself.
  • Nothing copied or stored from your systems. Answers are read live from your systems. Only if your organization chooses Saved answers, an optional add-on, are a person's own recent answers kept for a short time, in memory and encrypted. Results are held briefly and then deleted. What is kept longer is listed below.
  • Each person's own access. People and their agents act as themselves, so each system's own permissions decide what they can see and do.
Seven layers

Every request passes through seven layers before it reaches your systems.

Your AIany AI app that supports MCP, the SourceLace app's assistant, and AI agents
  1. 1
    Each person's own sign-inPeople connect each system with their own account wherever the system offers it, so its own permissions and sharing rules decide what they see. Connectors that use one shared account are labelled so, and an admin adds them knowingly.
  2. 2
    Access by groupAdmins choose which groups may use each system and object, and which may change them. Groups can come from your identity provider through single sign-on or SCIM provisioning.
  3. 3
    Data protectionHide sensitive fields so the AI never gets them, or mask them. Email addresses, phone numbers, US Social Security numbers, payment card numbers, IBANs and your own patterns are found automatically, in any text. The AI sees only placeholders (photos are the exception, and only if an admin allows them); the real values are shown only to the people you allow, in the SourceLace app and in the changes they confirm.
  4. 4
    Changes only with a person's yesRead-only by default. Where an admin allows changes, each one is previewed field by field and sent only after a person confirms it.
  5. 5
    Nothing copiedQuery results are held for at most 30 minutes, encrypted with your organization's own key when stored, then deleted. Your admin can make that shorter. Saved answers, only if your organization chooses them, at most 24 hours.
  6. 6
    Tamper-evident audit trailEvery call, by every person, app and agent, recorded with who, what and when, never the data itself, and chained so a changed or deleted entry shows. SourceLace checks the audit trail automatically every day and alerts your admins if anything does not match. You can also check it yourself at any time and export it with its proof, as CSV or JSON for a security review, or stream it to your SIEM or storage as it is written (Business and Enterprise).
  7. 7
    Works with the AI you chooseBring any AI app and SourceLace sends nothing to an AI model itself. In the SourceLace app, use the AI included in your plan or your organization's own AI key. The same layers apply either way.
Your systemsCRM, ERP, HR, finance, service desks, data platforms, databases, files and chat, read live
Data protection

You see the real values. The AI sees placeholders.

Choose what the AI may see of your data. Sensitive values such as email addresses and card numbers reach the AI as short placeholders, and hidden fields never reach it at all. The person who asked sees the real values in SourceLace, and anything the AI drafts or proposes is made with the real values only after that person confirms the preview. You decide which groups may see real values, and you can test a sample before you turn a rule on.

What SourceLace keeps

Your records aren't copied. Here is everything that is kept.

Query results are held for at most 30 minutes and never written to a database or a log. Files are read on demand and never indexed. A few things last longer, all encrypted with a key unique to your organization.

  • Saved chats and the files made in them. Chats in the SourceLace app are saved with the rows they show, and deleted after your retention period (30 days after last use by default).
  • AI agents' run summaries and pending changes. Changes waiting for approval include the before and after values. A run's progress is kept only until the run ends. Deleted with the agent and after your retention period.
  • Saved answers, only if you choose them. An optional add-on: each person's own recent answers from the systems your admin turns it on for, in memory only, for at most the time your admin sets (up to 24 hours), and never shown to anyone else.
  • Sign-ins and secrets. Each person's sign-ins to your systems and your admins' secrets, including your own AI key. Never logged or shown again.
  • Your setup. Skills, agent settings, groups and rules stay until you delete them. The audit trail holds who did what, never the rows returned or the values written.
Identity

People who leave lose access at once.

Sign in with Google, Microsoft or your own single sign-on, and require it for your domains. With SCIM provisioning, your identity provider adds people, keeps their groups up to date and deactivates them when they leave.

  • Everything ends together. Removing someone ends their sessions and AI app connections, and removes their sign-ins to your systems. Their AI agents pause for an admin to decide on.
  • No passwords of our own. For most systems people sign in on the system's own page, so SourceLace never sees their passwords. Multi-factor authentication is whatever your identity provider requires.
Security questions

Answers for your security reviewers.

Short, factual answers, grouped the way security questionnaires (such as SIG Lite, CAIQ and the SOC 2 criteria) ask them. Each one says what SourceLace does today, says plainly what is not in place yet, and links to the full answer in the docs.

Company and certifications

Does SourceLace hold a security certification such as SOC 2 or ISO 27001?

Not yet. SourceLace holds no certification or attestation today: no SOC 2 report, no ISO 27001 certificate and no HIPAA attestation. A SOC 2 Type 2 examination is in progress. Please do not record one in your vendor register. Details

Is SourceLace working towards SOC 2?

Yes. SourceLace has engaged an independent CPA firm for a SOC 2 Type 2 examination covering Security and Privacy. The observation period began on 1 November 2026, and we expect the report in 2027. SourceLace does not hold a SOC 2 report yet. Many of the controls an audit checks are already in place and are described on this page; the policies a SOC 2 audit expects are written and approved, while staff background checks and a penetration test by an outside firm are not yet. Details

Do you sign a DPA or a HIPAA BAA?

Not yet. A GDPR data processing agreement and a HIPAA Business Associate Agreement are not yet available, so do not connect systems holding patient data. Details

Hosting and data location

Where does SourceLace run, and where is our data?

In the United States only, with a major cloud provider named in our privacy policy. Backup copies are kept in a second US region. There is no EU or other regional hosting yet. Details

How is the hosting environment protected?

A private network where the database cannot be reached from the internet, a web application firewall in front, HTTPS only, threat detection and configuration checks that alert staff within minutes, and a record of every administrative action kept for at least a year. SourceLace calls your systems from fixed addresses you can allow in your firewall. Details

Can we keep everything in our own network?

SourceLace is a cloud service. To keep chosen AI agents' runs inside your network, use a SourceLace Runner: on the Enterprise plan only, on request. You choose for each agent whether it runs in SourceLace's cloud or on your runner. Your admins download it in the app: sign in to download. Runners and cloud and runners together

What does a SourceLace Runner need from our network, and what does it send to SourceLace?

A runner connects out only, over HTTPS on port 443, to SourceLace, your AI provider and your own systems. Nothing on the internet connects in, and SourceLace never connects to it. It sends SourceLace each run's status, steps and counts, never your data, and keeps everything else encrypted with a key only your organization holds. See the network a runner needs and Runner security.

Your data

Does SourceLace copy our data into its own store?

No. SourceLace reads your systems live. Query results are held for at most 30 minutes, then deleted, and files are read on demand and never indexed. The longer-lived exceptions are saved chats and the files made in them, and AI agents' run summaries and pending changes, all encrypted with your organization's own key and deleted after your retention period. Everything that is kept

Will people see more than they can see today?

No. With personal sign-in, every call runs as the person, so the system's own permissions apply. Your admins can narrow that further with access by group, read rules (which rows each group sees) and data protection rules. A few connectors can only use one shared account; they are labelled so, and an admin decides whether to add them, who may use them and which rows each group sees.

Can the AI change records in our systems?

Only where an admin has turned changes on for named objects of a system. Every change is previewed and sent only after a person confirms it, and previews expire within 10 minutes. SourceLace never sends email: mail connectors only make drafts.

Is our data separated from other customers' data?

Yes. Everything SourceLace keeps for your organization is tied to it, every request is checked against the signed-in person's organization, and your organization's secrets, chats and agent data are encrypted with your organization's own key.

Encryption and keys

Is data encrypted in transit and at rest?

Yes. HTTPS only, with TLS 1.2 or later, and TLS with certificate checks to your systems. At rest, sign-ins, secrets, saved chats, agent data and held results are encrypted with a key unique to your organization, and everything SourceLace stores, backups included, is also encrypted by the hosting environment. Details

Can we bring our own encryption key?

Not yet. Each organization's data is already encrypted with its own key, and you can bring your own AI key. Details

Identity and access

How do people sign in? Do you support SSO, SCIM and MFA?

With Google, Microsoft or your own single sign-on (OpenID Connect on Business and Enterprise, SAML 2.0 on Enterprise), and you can require it for your domains. SCIM provisioning adds, updates and removes people from your identity provider. SourceLace has no passwords of its own, so multi-factor authentication is whatever your identity provider requires. Single sign-on

Can SourceLace staff see our data?

SourceLace's operators manage your account, such as your plan and admins, on a page that never shows data from your systems, sign-ins, secrets or AI keys. Every change they make to your organization is recorded in your own audit trail under their name, and so is each time they read or export your audit trail. Your admins can list all of it on their own, under SourceLace staff on the Audit log page. Only a few named staff can reach the hosting environment, each through single sign-on with multi-factor authentication, and every action there is logged. Details

AI and model providers

Which AI model sees our data? Is it used for training?

With your own AI app, the model you chose there; SourceLace sends nothing to an AI model itself. In the SourceLace app, the AI included in your plan or your organization's own AI key. SourceLace does not train AI models on your data and does not send it to any AI provider other than the one your admin chose. Hidden fields never reach the AI, and masked values reach it only as placeholders. Details

Can someone talk the AI into revealing secrets or another organization's data?

No. The AI never holds a secret, organizations are kept apart by SourceLace's own checks on every request, and text inside your data cannot apply a change: every change waits for a person. AI safety

Audit trail and logging

What is recorded, and can we export it?

Every call, sign-in, settings change and agent run, with who, what and when, never the data itself. The trail is chained so a changed or deleted entry shows, checked automatically every day, and exportable with its proof as CSV or JSON. Changes SourceLace's own staff make to your organization are in it too, under their name. Failed sign-in and token attempts are recorded as well, including ones that cannot be tied to an organization, without the address or token itself. On Business and Enterprise it can also be sent to your own security tools as it is written, see the next question. Details

Send the audit trail to your own security tools

Every call is recorded, with who, what and when, never the data itself, and the record is chained so a changed or deleted entry shows. On the Business and Enterprise plans, SourceLace can send each new entry to your own tools within seconds: a signed webhook, Splunk, or an Amazon, Google or Azure storage bucket. Each batch carries proof, so your tools can check that nothing is missing. If a destination is down, SourceLace keeps trying and never skips an entry. Details

Retention and deletion

How long is data kept, and what happens when we leave?

Query results: at most 30 minutes, and your admin can make it shorter. Saved chats and agent runs: 30 days after last use by default, set by your admin. When you close your account, everything SourceLace stores for your organization, including its encryption key, is deleted 7 days later. The locked daily audit records (400 days), the records of requested web addresses (365 days) and backup copies follow their own schedules; see the privacy policy. Details

Backups and business continuity

How is SourceLace backed up? What are your RTO and RPO?

The database is backed up continuously and can be restored to any moment in the last 35 days. Backups are encrypted, with copies in a second US region. Restore drills are run once a year; the first was done in November 2026 and the restore worked. Recovery time and recovery point objectives are not yet defined in writing, and there is no uptime commitment or public status page yet. Your records stay in your own systems throughout. Details

Incident response and breach notification

Do you have an incident response plan? Will you tell us about a breach?

Yes. A written process is in use: four severity levels, one person leading each incident, harm stopped first, your admins told by email when your organization is affected, regular updates, and a written summary after a serious incident. If your data may have been reached, you are told without undue delay and as the law requires. A contractual notification time is not yet formalised. Report a suspected incident to security@sourcelace.com. Details

Secure development and vulnerabilities

How are changes checked, and how are vulnerabilities handled?

Every change goes through a pull request with automated checks, including security tests and checks against public databases of known vulnerabilities; a known high or critical issue blocks it. Server images are scanned, the hosting environment is defined in versioned files, and a failed release rolls back on its own. There has been no independent penetration test yet, and there is no bug bounty. Report a vulnerability to security@sourcelace.com. Details

Subprocessors

Who processes our data?

The cloud hosting provider, Anthropic for the AI included in your plan, the AI provider of your own key if your admin enters one, your sign-in provider, and Stripe for billing (never data from your systems). The list is updated before a new provider handles customer data. Privacy policy · What each one receives

Security review pack

Email security@sourcelace.com for any of these: the security questions in the docs (most questionnaires are answered there); the subprocessor list; a summary of how incidents are handled and reported to you; your own questionnaire, filled in by SourceLace, or a review call; and the security review pack, with more technical detail on encryption, key handling and hosting, shared under a non-disclosure agreement. Your admins can also export your own audit trail with its proof at any time. There is no SOC 2 report or penetration test report to share yet. What you can request

Found a vulnerability? Report it here.

More for your security reviewers: Security and privacy · Security questions · Data protection · FAQ · security@sourcelace.com