Frequently asked questions
Short, direct answers to the questions security, IT, procurement and business people ask about SourceLace, by topic, each with a link to the full page.
SourceLace is the AI-native agentic platform for enterprise systems. One agent across all your systems, working with the agents you already have, acting as each person, with every action approved and on the record. Every answer here describes what SourceLace does today. Where something is not in place yet, the answer says so.
- If you run business systems, start with Getting started and Connecting your systems.
- If you run revenue, customer success or finance operations, or an IT service desk, start with Asking questions and getting answers, AI agents and Changes and approvals.
- If you approve tools for security or IT, start with Security and data protection, Audit and logging and AI models and keys.
Categories
- Getting started: what SourceLace is, your first steps in the SourceLace app, and which other AI apps can reach it.
- Connecting your systems: which systems connect, who sets up what, and what your network needs.
- Asking questions and getting answers: what people can ask, how big an answer can be, and how to trust it.
- Changes and approvals: when SourceLace may change a record, and how every change is confirmed.
- AI agents: jobs that run on their own, what they may do, and how you keep them in check.
- Security and data protection: access, encryption, separation between organizations, and hiding or masking fields from the AI.
- Audit and logging: what the audit trail records, what it never records, who can read it, and how long it is kept.
- Governance and admin controls: roles, access by group, stop switches, limits and the controls admins have.
- Sign-in and single sign-on: how people sign in, single sign-on, automatic provisioning and sessions.
- Data storage, retention and location: what SourceLace keeps, for how long, where, and what happens when you leave.
- AI models and keys: which AI sees what, using your own AI key, and training.
- Compliance and contracts: certifications, agreements, questionnaires and incident notice.
- Plans and billing: what the plans include, pilots, seats, how AI use is counted, seeing and limiting AI spend, engine rows, when your contract renews, and whether unused credits roll over.
- The SourceLace Runner: running your AI agents inside your own network, on the Enterprise plan.
- Troubleshooting: common messages and what to do about them.
Can't find your question?
Search the FAQ with the box above, or search the whole manual with the box at the top of the page. You can also ask the assistant in the SourceLace app, or your own AI app: they search this manual and answer with a link. Write to support@sourcelace.com, or, for security questions, security@sourcelace.com.
All questions
Getting started
- What is SourceLace?
- How is SourceLace different?
- Which AI apps work with SourceLace?
- Can we use SourceLace from Microsoft Teams?
- Can we use SourceLace from Slack?
- How do we know what to set up next?
- Do I need the SourceLace app if I already use an AI app?
- How do I start?
- How long does it take to set up?
- What are skills?
- What is the difference between instructions and skills?
- Can I see what changed in a skill, or go back to an earlier version?
- What are projects?
- How do I pin a project, and who sees my pins?
- How do I start a chat in a project?
- How do I star a chat or move it to a project?
- Where are my chats in the sidebar?
- Can I ask SourceLace how to set something up?
- Where do I get help?
- Where do I see what changed in SourceLace?
- How much notice do we get before something is removed?
- How do I change my name, picture or dark mode?
- Does SourceLace use my Google picture?
- Where do I tell the assistant how to answer?
- Can the assistant help me learn SourceLace itself?
Connecting your systems
- Which systems can SourceLace connect to?
- Who sets up each system?
- Does adding a source give everyone access to its data?
- Can we connect the same system more than once?
- Can I use two mailboxes, such as a shared inbox and my own?
- What is the redirect URL?
- What does our network need?
- Can SourceLace reach systems in our data center?
- Can we connect a system SourceLace has no connector for?
- Can we use a vendor's own MCP connection inside SourceLace?
- What happens when a vendor changes one of its MCP tools?
- Should we use a vendor's MCP connection or SourceLace's own connector?
- Should a bucket be searched by names only or by file contents?
- Can we connect our own search index?
- Can I use a Google Sheet?
- What does our Salesforce admin do to connect Salesforce?
- Can we use our own Salesforce app instead of installing SourceLace's?
- What does Data 360 need besides Salesforce?
- Can we connect Salesforce Marketing Cloud?
- Can I combine Marketing Cloud with Salesforce in one question?
- What does our database administrator do to connect a database?
- How is a MongoDB Atlas cluster connected?
- Can we connect our own S/4HANA or SuccessFactors?
- Why can't SourceLace use one of our S/4HANA Cloud APIs?
- Which SAP interface types can SourceLace use with S/4HANA Cloud?
- What do our SAP teams do to connect SAP ECC?
- What does our Datasphere admin do to connect SAP Datasphere?
- What does our Snowflake admin do to connect Snowflake?
- What does our Google Cloud admin do to connect BigQuery?
- What does our Databricks admin do to connect Databricks?
- Does connecting Snowflake or Databricks cost anything when nobody is asking questions?
- What do our Amazon admins do to connect Amazon Redshift?
- What do our Amazon admins do to connect Amazon Athena?
- What do our Microsoft admins do to connect Fabric or Synapse?
- What does our HubSpot admin do to connect HubSpot?
- What do our Microsoft admins do to connect Dynamics 365?
- What do our admins do to connect Dynamics 365 Finance and Supply Chain?
- What does our NetSuite administrator do to connect NetSuite?
- What do we need to connect Gainsight or ChurnZero?
- Can Gainsight or ChurnZero be joined with our CRM in one question?
- Can Flexport, Finance and Supply Chain or remittance advice be joined with our other systems in one question?
- What do we need to connect Flexport?
- Who in Gong can connect Gong?
- What does our Oracle admin do to connect Oracle Fusion?
- What does our Workday admin do to connect Workday?
- What does our Google Workspace admin do to connect Google Drive or Google Sheets?
- What does our Amazon cloud admin do to connect Amazon S3?
- What does our Azure admin do to connect Azure Storage?
- What does our Google Cloud admin do to connect Cloud Storage?
- What does our Google Workspace admin do to connect Gmail and Google Calendar?
- What does our Microsoft admin do to connect Outlook mail and calendar?
- What does our Slack admin do to connect Slack?
- Why does Teams need a Microsoft admin?
- What does our ServiceNow admin do to connect ServiceNow?
- What does our Zendesk admin do to connect Zendesk?
- What does our Atlassian admin do to connect Jira?
- What does our Trino or Starburst admin do to connect Trino?
- What does our Amazon cloud admin do to connect lake tables?
- What does our EHR team do to connect health records?
- Do we have to register an app with Epic or Oracle Health ourselves?
- What goes in the FHIR source's Scopes field, and where do I pick SMART v1 or v2?
- Do the healthcare pack's agents work with Oracle Health?
- A health records search says it "took too long". What do I do?
- What do we need to connect a healthcare clearinghouse such as Stedi, Availity or Optum?
- What do our developers do to run a custom connector?
- Can a system we connected with the Connector SDK be joined with our other systems?
- What does the system's admin do to connect a REST or SOAP API?
- How do I make a long list from a REST or SOAP system, or a vendor's tools, faster in a question across systems?
- What does the vendor's admin do for a vendor's MCP connection?
- What does our search admin do to connect Azure AI Search, Elasticsearch or OpenSearch?
- What do Verified, Available and Pilot mean?
- Do the logos mean those vendors endorse SourceLace?
- Why does the list of sources show only some of them?
- Why does a connection have no Disconnect, or no Connect?
Asking questions and getting answers
- What can I ask?
- Why can't I see a record I know exists?
- Can I ask one question that spans several systems?
- Can I combine health records with HR or billing data in one question?
- How does SourceLace know two records are the same?
- Why did an answer say some records had no match?
- What happens if there's no matching rule?
- Does matching merge records or change my systems?
- Can I match by similar names (fuzzy matching)?
- Who decides how records are matched?
- Can I match differently from my organization?
- How do I propose a matching rule?
- Can I see which system and field each number came from?
- Why does a cell show a dash instead of being blank?
- How do I see which record in my system a row came from?
- Does SourceLace remember which records it matched?
- What does Check mean in an answer?
- Can I search all our systems at once?
- Why did the search leave out one of our systems?
- Can SourceLace answer a repeated question without asking our system again?
- How many rows can a question return?
- How do I know an answer is complete?
- What is the line under an answer?
- Why does it say "7 of these 10 rows have no match in Snowflake"?
- Can I see just the answer, without every table the AI read?
- Where did the list of steps go?
- Can the AI make up numbers?
- The assistant says a field does not exist, but I know it does
- Can I make charts and documents?
- Can SourceLace read our files?
- Can SourceLace search inside my files?
- How do I find files faster in S3, Azure Storage or Google Cloud Storage?
- Can we teach SourceLace our own business terms?
- Can I search my old chats?
- Why are big questions fast?
- Do I have to switch anything on for my database to work out totals?
- Why does a total in SourceLace have more groups than my database's own report?
- Why do "West" and "WEST" show as one group in a total?
- Can SourceLace group text the way my own database does?
- Why does a step say "Protected field left out"?
- Can I make the columns of an answer's table wider or move them?
- Can I make one table or chart wider?
- How do I know how full a chat is, and what happens when it is full?
- Can I see when a message was sent?
- An answer failed. What can I do?
- Why does a table say it was built by SourceLace from a plan?
- Does a file say what SourceLace checked?
- Can I export the rows of a table to CSV or Excel?
- Why does sorting only change the rows I see?
- Can I attach a photo to a question?
- Why does a step say it was corrected?
- What does Redo on all rows do?
- Can I choose whether my next question uses the data already read?
- What does "Answered from" mean in the answer check?
- Why is a table folded under "Built from"?
- How do I search or sort a list?
Changes and approvals
- Can SourceLace change records?
- Is SourceLace read-only by default?
- How do I confirm a change?
- What if someone changed the record after my preview?
- Can SourceLace send email?
- Can SourceLace put a meeting on my calendar or invite people?
- Can SourceLace delete records?
- Who approves changes that an AI agent proposes?
- Who is told when an agent's change is waiting for approval?
- What happens if nobody approves an agent's change?
- Can I approve several changes at once?
- Can we require a second person, or only approve large changes?
- Can we stop the AI from setting some fields?
- Can an AI agent change records without approval?
- Can a group have read access but not change access?
- What can a vendor's MCP tools change?
AI agents
- What is an AI agent?
- What is Agent Architect, and what does it send to the AI?
- Can different teams use Agent Architect on different data?
- What does the conversation before a run cost?
- Can an agent use our own deck template, playbook or past reports?
- Can I upload a file instead of linking it?
- Does Agent Architect make up how much time an agent saves?
- Can I design one agent for my whole team?
- Can our admin see how much each person uses Agent Architect?
- How do I find an agent's runs from a certain day or hour?
- Why does a run say "Waiting for approval" or "Lost"?
- How do we avoid costly mistakes with agents?
- Which plans include agents?
- Can an agent see or do more than its owner?
- Which systems can send events to SourceLace?
- What is the difference between a webhook and A2A?
- Can a record or an email trick an agent into doing something else?
- How do we try an agent safely before it goes live?
- What can start an agent?
- Can an agent start before a renewal date, or when something has waited too long?
- Can an agent start when other agents have finished?
- Can an agent ask our finance team to approve a discount before it goes ahead?
- Can every agent's results read the same way?
- Can our agents ask each other for help?
- Which systems can start an agent?
- Can one agent cover several Salesforce orgs or SAP systems?
- Can an industry pack set up agents for several of our systems, such as a Salesforce org for each region?
- Can one agent send to several places, such as an email and a Slack message, or update every org?
- Can an agent email or post the report it made, or do it without me approving each one?
- Can an agent save its reports to SharePoint, OneDrive or Google Drive?
- Can I start an agent from Slack or Teams?
- Can I start an agent by sending it an email?
- How do we stop an agent?
- What is kept from an agent run?
- Can our agents run inside our own network?
- Can agents start when something happens in Salesforce?
- Can agents start from JMS, IBM MQ or Kafka?
- Can agents start when data changes in MongoDB, in Gmail or in our own system?
- How do I know an agent is ready?
- What happens when I rate an answer?
- Can an admin give an agent to someone else?
- Can an admin start someone else's agent from another AI agent?
- How do I know whether my agents are doing useful work?
- Why does a run's journey show mostly waiting?
- Can I change the columns of the agents list?
Security and data protection
- How does SourceLace keep people to their own access?
- Does Search everything copy or index our systems?
- Is our data encrypted?
- Can we bring our own encryption key?
- Is our data separated from other organizations?
- Can SourceLace staff see our data?
- What happens to the secrets and keys we enter?
- Can we hide or mask fields from the AI?
- Do the data protection rules apply everywhere?
- Can SourceLace be tricked into running a harmful query?
- How do we report a security issue?
- Is it safe to add a vendor's MCP connection?
- Where can our security reviewers get more detail?
- How is SourceLace's hosting protected?
- What does "shown to you only" mean in an answer?
Audit and logging
- What does the audit trail record?
- What is never in the audit trail?
- Who can see what my people asked?
- Is everything audited, from every AI app?
- Where do admins read the audit trail?
- Is the audit trail tamper-evident?
- How long is the audit trail kept?
- Can we export the audit trail or send it to our SIEM?
- Are SourceLace staff actions recorded?
- Can we prove that results were deleted?
- Are AI agents audited?
- Are data protection rules and masking audited?
- Are sign-in and provisioning changes audited?
- Does SourceLace record our AI use?
- Do SourceLace's own system logs contain our data?
- How do I look at what happened in one hour of one day?
Governance and admin controls
- What roles are there?
- How do we control who can sign in?
- Can I limit a source to some people?
- Can access by group give someone more than their own access?
- Can people add their own accounts, and can we control it?
- Can I limit which rows a group sees?
- How do we remove someone?
- Can we act quickly in an incident?
- Who decides what the AI may see?
- Where did Organization settings go?
- What is the Health page?
- Can we control who creates AI agents?
- Can we set stricter limits than SourceLace allows?
- How do we know our setup is sound?
- Can we copy a setup from a test organization?
- Which account does the setup assistant read my system with?
- Will the setup assistant keep suggesting what I rejected?
- Can SourceLace change our settings without us?
- Can admins manage SourceLace from chat?
- Does SourceLace learn from what our people ask?
- How do I let people attach photos in the chat?
Sign-in and single sign-on
- How do people sign in to SourceLace?
- Which identity providers do you support?
- Can we require single sign-on?
- How do we switch to another identity provider?
- Do you support multi-factor authentication?
- Can our identity provider add and remove people automatically?
- Can groups come from our identity provider?
- How long do sessions last?
- What if our identity provider is down while single sign-on is required?
- Can one person belong to two organizations?
- Why does SourceLace ask me to Allow when I connect my AI app?
Data storage, retention and location
- Does SourceLace store our data?
- Where is SourceLace hosted?
- Can we make retention shorter?
- How long are chats and agent runs kept?
- Where are the files my chats and agents made, and when are they removed?
- Are our files copied or indexed?
- Are our records in SourceLace's backups?
- What happens to our data when we leave?
- Can a person have their own data deleted?
- Does SourceLace keep the names of Jira people?
AI models and keys
- Does SourceLace send our data to an AI model?
- Does the AI see our sensitive data?
- Which AI does the assistant in the SourceLace app use?
- Can each person choose their own AI model?
- Does SourceLace summarize our chats, and can we stop it?
- Can we use our own AI key?
- Can the AI run in our own Azure or Amazon cloud account?
- How is our AI key protected?
- Does SourceLace train AI models on our data?
- Does using our own AI app cost SourceLace AI credits?
- What does SourceLace's own AI see when it helps with setup?
Compliance and contracts
- Is SourceLace SOC 2 certified? Do you sign a DPA?
- Do you have written security policies and staff checks?
- What can we request for a security review?
- We have a security questionnaire to fill in. Where do we start?
- Has SourceLace had a penetration test? Is there a bug bounty?
- Can we use SourceLace with patient data?
- Who are your subprocessors?
- Will you tell us before adding a subprocessor?
- Will you tell us about a breach?
- Do you have an incident response plan?
- Is there an uptime SLA or a status page?
- Can our contract differ from the standard terms?
Plans and billing
- What plans are there?
- Can we try SourceLace before we sign a contract?
- When does my contract renew?
- Do unused credits roll over?
- How is AI use counted?
- How do I see and limit AI spend?
- Do large questions across systems cost extra?
- What is a seat?
- Which connectors does each plan include?
- How do we pay?
- What happens if a payment fails?
- Are features we don't have hidden?
The SourceLace Runner
- Can we keep our data in our own network?
- Who can have a runner, and what does it cost?
- What happens to our runners if our contract changes?
- How do we get the runner?
- How long is a runner version supported?
- When do our runner licenses end?
- Before we choose a runner, what should we weigh?
- What does a runner need, and where can it run?
- Which network connections and ports does a runner need?
- How can our security reviewers check a runner?
- Can our agents run inside our own network?
- Can we run some agents in the cloud and others on our own runner?
- Do our data protection and read rules apply on a runner?
- Does a runner need an inbound port?
- What happens to a run if the runner restarts or loses its connection?
- Can a runner keep working if one of our machines fails?
- Who is responsible for what with a runner?
- What do we send SourceLace support about a runner?
Troubleshooting
- Why does a source say "not connected" or "needs sign-in"?
- Why is a system "not set up on SourceLace yet" or "not available yet"?
- It says a label is the label of two sources.
- I am told I am not allowed to sign in. What now?
- It says my groups do not give me access.
- How do I check that my connections and the AI assistant are working?
- How do I find out which of our connections or agents is failing?
- It says a result has expired.
- It says a change has expired or was already applied.
- The assistant in the app stopped answering. Why?
- One chat says the AI model "could not continue this chat", but new chats work.
- It says a system or the AI is busy: "try again in N seconds".
- A system could not be reached or did not answer in time.
- Single sign-on sign-in fails.
- An AI agent run failed.
- Data 360 sign-in fails with OAUTH_APPROVAL_ERROR_GENERIC, but Salesforce works.
- Salesforce sign-in fails with OAUTH_APPROVAL_ERROR_GENERIC or "user is not admin approved".
- A database "could not be reached".
- A database does not accept my username and password.
- SAP says it refused my new sign-in, and connecting again does not help.
- Snowflake says it has no warehouse to run my question on.