Security
The controls that protect research done through Passim, stated specifically, including what is not yet in place.
Last updated: September 13, 2026
Passim is built to be used inside a law practice. This page states the controls that protect research done through it, in specific terms. Passim is not yet independently audited; where a control is planned rather than in place, this page says so.
At a glance
What Passim receives, and what it doesn't
Passim is a retrieval service. Your AI assistant sends it a research query over MCP and gets back legal sources, citations, and research signals. Passim never receives the rest of your conversation, your documents, or your client files. The assistant is a separate system under its own agreement: it decides when to call Passim, keeps its own copy of the results under your workspace's settings, and is not a Passim subprocessor. Passim cannot configure its retention, training, or administrator controls, and does not claim to.
No training on customer data
Passim does not use account data, queries, tool calls, saved work, or support messages to train or fine-tune any model, and does not send your queries to any AI provider. AI is used on public legal materials only, to summarize and classify them.
Queries are processed, not stored
The text of a query is used to answer it and is not written to logs or usage records. Usage records keep the calling identity, tool name, date, counts, timing, and error or rate-limit totals. Queries travel in request bodies rather than URLs, so they never appear in access logs; error monitoring is configured not to capture request bodies; and when an operational problem needs correlation, content fields are fingerprinted rather than kept. Operational logs are deleted after 90 days, and that deletion runs as a scheduled control with its own evidence.
Encryption and network
Traffic to the website, the API, and the MCP server is encrypted with TLS. The production database runs on managed infrastructure and is reachable only over private networking, never from the public internet. The application and scheduled jobs use separate least-privilege database roles, and secrets live in the platform's secret store rather than in code.
Access control
Each person signs in with their own identity through OAuth. There are no shared keys. Issued API keys are stored as hashes and can be scoped, given an expiry, and revoked one at a time. Administrator access requires multi-factor authentication, and access to every provider account is reviewed at least quarterly and removed when it is no longer needed. Browser sessions are protected by a same-origin check on every change.
Backups and recovery
The production database runs with high availability and point-in-time recovery. Recovery has been exercised, not just configured: an isolated restore of a production recovery point was verified against the schema and the corpus, a rollback-only write was tested, and the restore target was discarded, with production unaffected throughout. Restore tests recur on a set schedule.
Secure development
Every release passes automated tests, dependency and static-analysis checks, and secret scanning. Dependencies and CI actions are pinned to reviewed versions, production installs only from the reviewed lockfile, and reviewed findings are held to an exact baseline so nothing new slips in unexamined. Vulnerabilities are fixed on fixed timelines: critical within 7 days, high within 14, medium within 30, and low within 90.
Monitoring and incident response
Security-relevant operating evidence is collected automatically every day and kept for 90 days, with privacy and size gates on what is collected. Incidents are triaged by severity, contained, investigated, and documented. If an incident affects you, we notify you without unreasonable delay, with a target of 72 hours from confirmation.
People
Passim is operated by its founder, a practicing immigration attorney. Production access is limited to the founder, every provider account uses multi-factor authentication and a unique identity, and no employee, contractor, or automated agent has standing access to customer data.
Vendors
Every provider that processes customer data on Passim's behalf is listed on the subprocessors page with its role and location. Each is reviewed before use and at least annually, and the list is updated before a new provider begins processing customer data.
Domain and email
The production domain is protected by DNSSEC, and a strict DMARC policy instructs mail receivers to reject unauthenticated mail claiming to come from it.
Independent assurance
Passim is not SOC 2 certified. Its control set and evidence are being prepared for a SOC 2 Type I readiness assessment, and independent penetration testing is planned and has not yet been completed. This page will say so when either is done; until then, the controls above are Passim's own statements and are available for review in security questionnaires on request.
Reporting
To report a vulnerability, email chris@chrishammondlaw.com with "Passim security report" in the subject line, and please do not post exploit details publicly. To ask for review of a public record in the library, send the source URL, the document identifier, and the concern to the same address.