Searchable field-level encryption on Supabase with CipherStash
supabase.com
supabase.com
If you can run “select * where secret_col == 10”… why does it matter that the column is encrypted?
If a column is encrypted using standard encryption (like AES-GCM) then the values are non-deterministic and fully randomized. That means that if you encrypt the same value twice, you'll get 2 different ciphertexts.
So the query: select * where secret_col == 10
Would actualy be: select * where secret_col == encrypt_aes(10);
And values in secret_col will never match (because the output of encrypt_aes will be different every time, even for the same input).
A common way around this is to use deterministic encryption which eliminates the randomization at the cost of a slightly weaker security model. What leaks is the ability to see if any 2 plaintexts are equal (because they have the same ciphertext) - what you actually want in the case of search.
You have to be careful implementing deterministic encryption though: don't use AES-GCM with a fixed nonce because the scheme completely breaks. You can use CBC mode but then you lose authenticity. We use AES-GCM-SIV (synthetic IV which retains authentication but is secure under a fixed nonce) and HMAC (keyed hashing).
But there are approaches to solving queries like: -- range/order SELECT * FROM foo WHERE x > 10; SELECT * FROM foo ORDER BY x; -- fuzzy text SELECT * FROM foo WHERE name ~ "dan";
These capabilities are all based on public research: For example, order/range uses: https://eprint.iacr.org/2016/612.pdf
Our docs are quite limited at the moment (fixing as quickly as we can!) but you can see the current list of supported queries here: https://cipherstash.com/docs/stack/cipherstash/encryption/qu...
What does the risk profile look like in a full leak of the encrypted database?
If you had a column of integers are you able to order them without decrypting? Check that values are the same? Identify null values?
We actually have a number of different schemes which trade leakage for performance/storage overhead/compatibility. Each one has its own type in Postgres so you just create a column with whatever type you want.
The weakest (most leakage) is OPE - order preserving encryption. We use an encoding scheme that works like scientific notation so an attacker with a DB snapshot would learn relative order but not the size of the gaps between values. Its major benefit is compatibility - it works everywhere.
The strongest is Block ORE - Order Revealing Encryption. A DB snapshot reveals nothing more than standard randomized encryption (IND-CPA2 semantic security). The tradeoff is that BlockORE values are much bigger: 32-bit integer goes to 384 bytes. On Postgres its still very fast and works with standard B-trees.
> If you had a column of integers are you able to order them without decrypting?
Yes. For either scheme, its just an EQL function:
SELECT * FROM foo ORDER by eql_v3.ord_term(x);
> Check that values are the same?
-- $1: encrypt(query) SELECT * FROM foo WHERE eql_v3.eq_term(x) = eql_v3.eq_term($1);
> Identify null values?
NULL is just like any value for OPE/ORE - encrypt it and use that to query -- $1: encrypt(NULL) SELECT * FROM foo WHERE eql_v3.eq_term(x) = eql_v3.eq_term($1);
NULL values can be encrypted with either scheme. Very safe to do so with the ORE scheme (fully randomized values). If encrypting strings, NULLs would reveal length but you can always pad if not leaking value length is important to your security model.
In the OPE scheme, NULL would encrypt deterministically so if you expect a lot of NULLs in your data this could leak the distribution.
We are working on a capability now called Leakage Tuned Domains. The goal here is not to eliminate leakage but scope it to a specific data type (e.g. birthdays) such that any leakage cannot be an advantage to an attacker. Think k-anonymisation or differential-privacy on steroids.
Some links if you want to learn more: https://github.com/cipherstash/encrypt-query-language https://cipherstash.com/blog/fixing-a-1-in-256-bug-in-cllw-o...
Any adversary who knows the distribution of the plaintext data can guess what your encrypted values are, with accuracy that improves as your sample size increases.
There were a bunch of papers on this 10-12 years ago. (Full disclosure: I was an author on some of them.)
This is all started when I was the CTO of a health-tech and the engineers all had access to patient data. They needed DB access (data migrations, managing backups, reading non-sensitive fields) but not access to the sensitive values. 10m+ patient records - it made me nervous.
Row-level Security (RLS) was a nightmare, it was slow, easy to mess up and could be overridden by DBAs anyway. Plus the application connected to the DB using a service account so RLS was useless for end-user access control.
CipherStash solves the following:
* Hide sensitive data from direct database accessors (authorized or not!)
My original use-case. Direct access to the DB doesn't reveal sensitive data (unless the user also has permission to decrypt a specific record).
* Support or even replace application access controls
Decryption happens in the application layer and is based on end-user identity. That means a gap in application defenses is mitigated by the encryption. If the user can't decrypt the value then they can't access the data.
* Access auditing
Every sensitive data access is recorded (and by who) which is very useful for compliance and incident response. SQL query auditing is great for knowing what queries were run but it doesn't tell you what data was actually returned. Its also blind once data actually leaves the DB and can inadvertently leak sensitive data itself. Using key accesses with non-identifying value IDs to audit decryptions makes the auditing more reliable and super granular.
We have customers who use the tech to prove to their users (often large enterprise) that no internal staff can access the data.
* Agent access to data
The encryption shifts access control decisions down to the data layer on a per-value basis. This is incredibly useful for gating access to agents that need access to data. Every decryption requires authentication (e.g via Supabase Auth, Auth0 etc) and combining it with agent identities (creds issued to agents) means you can not only limit what agents can access, you can also audit exactly what _was_ accessed.
The searchable encryption is the enabler: value level encryption is what makes all of this possible. Searchable encryption means you can have encryption and queries (from apps etc) still work for authorized users.
Hope that helps! I'm so in the weeds I don't know if I explain things that well sometimes!
If the value changes would you be able to get from the id you have audited back to the value that was returned at that point in time?
The answer to your question is yes.
Explanation: The identifier is actually for the key that encrypts the value (1 unique key per value).
1. When the value is encrypted for the first time, it gets an ID. 2. When its decrypted, the application requests the key for that ID from the key-server (key-server records that the data was accessed)* 3. When updating, the same data key is used so the ID is persistent*
* Technically the key server doesn't return a key, it generates partial key material that can be used to derive the data key in the app. It can do this at up to 10,000 keys per second. * You can also tag each value with the table/column name and the row-id to link everything together
The 3 values form the "descriptor" of a value: `table/column/id`.
It depends on your setup. If you're using Supabase, one way is to send the logs to Clickhouse and use the Clickhouse partner integration to query the audit logs and join it to the actual data.
Keeping only the ids in the audit log means you need to stitch the data together later. This means you don't accidentally leak data via your audit trail!
> SQL query auditing is great for knowing what queries were run but it doesn't tell you what data was actually returned
With what you’re saying - sounds like cipher doesnt tell you what data was actually returned either - it will only tell you if the user could have received a decrypted version of the data right?
Query logging would tell you what queries were executed. E.g the following would tell you that 10 user records were returned with name and email but not which users.
SELECT name, email FROM users LIMIT 10;
CipherStash can tell you the ID of every row that was returned, like:
users;email,name;1 users;email,name;2 users;email,name;3 etc
If you need to know the actual values instead of just the IDs (say when doing an investigation), you can do a query to join the audit data with the users table.
Assuming audit logs are in a table called audit with a column called record_id and a timestamp:
-- full forensic analysis SELECT name,email FROM audit LEFT JOIN users ON users.id = audit.record_id WHERE timestamp BETWEEN $1 AND $2;
And because the data is encrypted, that query will ALSO be audited. Not even admins escape the auditing :p
All of this is important for "materiality" investigations. If you suspect some data has been accessed inappropriately, knowing how much and exactly which records is very useful - especially if deciding if customer/regulator notification is required.
Keys are not stored in the database or in the application. Every data key is derived at query time via a 2-party system: 1. by the key server which manages root-key material (stored in an HSM or traditional KMS) 2. keys derived via a user credential at the time of the query
SQL Injection case: An adversary signs up to your app, pulls of a SQLi and you might think that the app will just decrypt the values. It won't because the adversary's auth cred can't derive keys to decrypt anything but the data they are explicitly allowed to access.
RCE/shell compromise: Because keys are derived on demand these kinds of attacks are largely inert. If the attacker could read memory from the running app on-demand then they could conceivably see credentials in flight but each key only ever decrypts one value so the blast radius is limited.
On the cost of decryption: Queries typically only return a subset of the values in a table and, importantly, don't need to decrypt anything at all. Querying and decryption are independent. If you DO need to decrypt a large volume of data, CipherStash can do it at 10,000 values per second.
Query performance is under 1ms for simple queries and no-more than 200-300ms for more complex queries.
Full benches here: https://github.com/cipherstash/benches
No security tool is perfect, nor "gold plated" but this tech meaningfully reduces blast area of attacks, makes many attacks inert and gives you an incredibly reliable audit trail.
I’m not a cryptographer and I can see flaws in your proposed system, I imagine someone qualified could poke some pretty glorious holes in your snake oil.
Now, if an attacker could gain access to the client key AND a valid JWT then they would be able to decrypt until the JWT expired. Blast radius reduced but we don't stop the attack entirely. An attacker with that much advantage is pretty hard to stop dead. But note that 2 things are required for the adversary to do this - where as in most systems, getting a JWT or a decryption key alone would be enough.
If you're not a cryptographer, then I'd suggest you reserve judgement about what flaws you think might be in the system. We have cryptographers on our team, have connections with universities in the US and Australia and our work is based on published, peer-reviewed papers. E.g. Lewi-Wu 2016 and Syalim et al 2011 and 2021.
What is the benefit of all this?
TBH, I'm so close to this now I forget that part isn't obvious! You literally can't do:
select * from blah where secret_col == 10
And get meaningful results. You can only do something like:
select * from blah where secret_col == Encrypt(10);
And in order to generate that encrypted query, you must be authorized. So an adversary can't simply do trial and error queries to guess records.
If not, then a lot of the data could be easily reconstructed.
Docs here: https://cipherstash.com/docs/stack/cipherstash/encryption/pr...
And a mention in our April update on Prisma Next: https://www.prisma.io/blog/prisma-next-roadmap-april-milesto...
If you want to give CipherStash on Supabase a try, using Prisma Next is the smoothest experience.
It is both remarkable and depressing how _little_ information is given, and how buried it is on the CipherStash website... _any_ information on what their security and/or threat model is, what is actually stored, how encryption and search works, or any trade-offs involved.
Just to list a few pages that tell you next to nothing:
https://cipherstash.com/docs/stack/reference/what-is-ciphers...
https://cipherstash.com/docs/stack/cipherstash/encryption/se...
https://cipherstash.com/docs/stack/cipherstash/encryption
I eventually found:
https://cipherstash.com/docs/stack/reference/security-archit...
Which... it sounds like 'searched without being decrypted' means... it encrypts your query against their fast KMS and uses that to compare against indexes that were also encrypted with the same KMS? And ORE/OPE is an optional mode when you want range support.
Queries are performed by generating SEM which is "compared" to values in the table and the source encrypted values are returned to the caller (usually an application or agent) which does the actual decryption.
This is crucial because if you have a table of millions of rows, you don't want to decrypt every one to find the results. SEM works with built in postgres indexes (B-tree, GIN) so a query over millions of rows executes in literally milliseconds. Then the resultset (typically only 10s or 100s or rows) can is all that needs to be actually decrypted.
You're right that we use OPE and ORE. OPE is necessary on Supabase right now due to some limitations on their side but that's being addressed. ORE is more secure and almost as fast but requires "CREATE OPERATOR FAMILY" permissions in postgres to work.
We also use encrypted bloom filters and structured encryption (STE) for queries over JSON objects.
This sounds like a back door. Is it?
To me, the whole article feels super-unclear about what exactly is involved.
Can HN folks who know more weigh in?
Our model is that data is encrypted in the application before being saved to the database. In order to encrypt, decrypt or query, you need to use the SDK (@cipherstash/stack). Connecting via psql, pgadmin etc, you'll only ever be able to see encrypted values (those tools don't know how to decrypt).
Proxy connects to the database and then you connect to the proxy (with psql or whatever). Proxy can now perform queries and decryptions on your behalf so that you can still access data if you need - hence "escape hatch".
The proxy authenticates using your credentials (or an access key) and interacts with the key management server.
Hope that helps!
The tech is the result of 6 years of work. We're a team of 8 engineers and have been working on it full time thanks to VC funding.
I'm obviously biased but its legit tech. We've invested in making it fast, secure and able to support most common query patterns.
Our github repos might be better than our website but I'll certainly admit that our docs/marketing needs some work!
Checkout out: https://github.com/cipherstash/stack https://github.com/cipherstash/encrypt-query-language https://github.com/cipherstash/proxy
Any chance you’re hiring for a related role? Technical marketing and technical documentation are my jam!
CipherStash isn't something that we just threw together on a weekend. We've spent years developing it. I've personally invested almost every waking moment for the last 6 years working on this (and the prior 2 preparing to quit my job so I could work on it full time).
I get that it seems like snake oil to some. But that's just because we don't do a good enough job of explaining what we've created. We're working on that - we're engineers, not marketers.
There have been prior attempts at making tech like this work, and maybe because those approaches were limited, its tainted how folks see encryption tech. We've studied many of the previous attempts and invested in the engineering required to overcome the problems.
Key management - we made a 2 party key system that is faster and has a better security model than AWS KMS alone. Not because we thought it was fun, not on a whim but because it was the only way to solve the real security challenges we faced when it comes to protecting data.
Searchable encryption - that is less secure than homomorphic encryption but FAST. Like under a few milliseconds for most queries (homomorphic would take literally minutes or worse). And a whole lot better than no encryption at all. It uses a key per value.
Encrypt query language - a framework that makes searchable encryption work in standard postgres. No exotic index schemes, no custom builds. Just SQL functions.
Typescript and Rust libraries to make this _almost_ drop-in for Prisma Next, supabase-js and Drizzle.
Identity integrations with Supabase, Clerk, Auth0, and Okta.
A postgres proxy that automatically encrypts and decrypts data for authorized requests - for when using the SDK isn't possible/convenient.
And an audit system that records every access and can send data to anything you can connect to AWS firehose or an S3 bucket.
This is real and IMHO is exceptional engineering...anyone know a great technical marketer?!
https://github.com/cipherstash/stack https://github.com/cipherstash/benches https://github.com/cipherstash/proxy https://github.com/cipherstash/encrypt-query-language
The premise first: organisations shouldn't have to make a mutually exclusive choice between encrypting their data and being able to query it. That's the reason CipherStash exists, and everything below follows from it.
The threat model in one sentence: anyone with access to the database itself - dumps, backups, stolen credentials, SQLi, a hosting provider, an over-curious DBA - should learn as little as possible about sensitive values, and every legitimate decryption should require authentication and leave an audit record. We're explicitly not claiming to defend against an attacker who fully owns your running application with valid credentials. Nothing at this layer can, and anyone who tells you otherwise is selling snake oil.
The components:
* ZeroKMS: the key service. Root key material sits here, HSM-backed. Every encrypted value gets its own unique data key, but ZeroKMS never has it - keys are derived in your application from a seed we return plus a client key that never leaves your infrastructure. Two parties are needed to derive any key, so CipherStash cannot decrypt your data. Not "won't", can't.
* CTS: the token service. Exchanges tokens from your identity provider (Supabase Auth, Auth0, etc.) so decryption is bound to end-user identity rather than a service account. This is what makes the SQLi case boring: an attacker's session can only derive keys for data that user was allowed to read anyway.
* Auditing: because every decryption requires a key derivation, the key service sees every access - which value (by key ID, not content) and by whom. Query logs tell you what was asked; this tells you what was actually decrypted.
There are two ways to integrate, and you can run both:
* At the application layer, via the ORM adapters (Supabase, Prisma, Drizzle). Encryption and decryption happen inside your application process - the database only ever stores ciphertext and encrypted index terms, and plaintext exists nowhere outside your app. This is the tightest trust boundary and the mode the identity-bound stuff above is built for.
* At the SQL layer, via the Proxy - a Rust Postgres proxy that runs in your infrastructure. It encrypts and decrypts transparently, before results reach the client, so anything that speaks Postgres works unmodified: psql, BI tools, migration scripts, legacy apps you can't touch. Crucially, credentials through the Proxy are still tied to an individual identity, not a machine identity. There are two layers of auth in play: a machine credential identifies the Proxy instance itself, and the user's identity is tunnelled through it - so the engineer running psql derives keys as themselves, subject to the same access policy and audit trail as everything else. The trade-off between the two modes is where plaintext appears, not who's accountable. Escape hatch, not back door - a back door would be a bypass of the access controls - this isn't.
Key rotation is handled by the same re-encryption scheme that powers derivation: root key material can be rotated server-side without re-encrypting the data itself, because the stored ciphertexts never depended on a long-lived key sitting next to them.
On leakage, since that's the crux of the scepticism: yes, searchable encryption leaks. Equality indexes (HMAC) reveal when two values match, so frequency. Range indexes (ORE, Lewi-Wu 2016) reveal ordering. Text indexes (encrypted bloom filters) reveal probabilistic token membership. This isn't a bug we're hiding, it's the fundamental trade: for the database to use an index it has to compare something, and whatever it compares, it learns. The schemes that leak nothing (FHE, ORAM) are bullshit slow. What we've done is make the leakage bounded, documented, and opt-in per column, per query type. If the frequency or ordering of a column is itself sensitive - salaries being the classic example - don't index it. Encrypt it plainly and filter after decryption in the app. Most columns need no search at all, and the default is exactly that: no index, no leakage beyond ciphertext length.
The docs criticism lands and we're fixing it - the security architecture page (https://cipherstash.com/docs/stack/reference/security-archit...) now covers the primitives, key hierarchy and per-index leakage in one place, benchmarks are at https://github.com/cipherstash/benches, and the crypto that matters is open source and auditable: https://github.com/cipherstash/encrypt-query-language, https://github.com/cipherstash/ore.rs, https://github.com/cipherstash/proxy.
> We're explicitly not claiming to defend against an attacker who fully owns your running application with valid credentials.
and dandraper says:
> I was CTO [...] and the engineers all had access to patient data. They needed DB access [...] but not access to the sensitive values > [...] > CipherStash solves the following: > * Hide sensitive data from direct database accessors
So putting these together is this implying that a developer who wants to test things on a large data set is given access to the database directly, but not access to debug the application? Surely for example if they could load the app in a debugger, step through it, and execute arbitrary code (eg. if they are developing on it) that includes "fully running the application"? So are you saying that scenario is not secure?
If so then I'm struggling to see how a developer would use this realistically. As a dev how am I supposed to work on an application without access to be able to change the code (assuming here then that "change the code" means I can also add a statement that dumps a decrypted value to the log)?
If the security level we want is something like "devs need to be able to test migrations work on the actual data" then why would I use this instead of just NOT giving direct db access and having an audited environment spin up a db copy and container and just output the logs of whether the migration worked, without access to query the db directly. If I do that then nobody needs direct db access right? But then why would I need CipherStash?
What CipherStash does is let you specify specific columns that you want to encrypt. The application (via our SDK), encrypts values for that field before it’s saved and optionally decrypts it again when it’s read. Decryption is performed when the user provides a valid credential (JWT).
This would allow you to give your devs full access to the production DB. You don’t have to give them permission to decrypt any or specific fields if they don’t need it.
So for your use case that would possibly save a lot of effort - no need to create a santized copy of the database (which in itself is fraught).
What James was referring to was the attacker threat model. Decryption happens in the application code. So an attacker who can steal encryption keys as well as be a valid JWT would be able to decrypt data for as long as the JWT is valid. That scenario is not something CipherStash can fully prevent.
What it does prevent is unauthorised access to data. Whether via the database directly or via the application.
What’s important here is that it’s the data itself that protected. If you move encrypted values from your prod DB to test or dev, the same rules apply. A user who can’t read “email” values that were encrypted in prod would still not be able to read those values if you moved them somewhere else.
Does that clarify things?
So maybe let me simplify my question: If I set up both a copy of a database with values encrypted by CipherStash, and an application server connected to it, and the application server has enough access to do queries on sensitive data (eg "WHERE sensitiveField = ?", pretend it needs that for a migration), and a developer has root access on the application box while it's running, and a gdb session connected to the app, can they see any sensitive data or no?
If they can't, then that's genuinely cool, and I probably need to look into the workings a bit more!
Having a copy of the production database and a local copy of the application and a debugger is insufficient to be able to encrypt/decrypt any values or run any queries involving encrypted search terms: you'd still need to authenticate with ZeroKMS.
Having root access to a production server running an application is a different matter because you'd have access to the entire address space of the app that real authenticated users are interacting with. A sufficiently motivated adversary would be able to intercept auth tokens allowing impersonation of an authorised user.
The application server process is a trust boundary in this model and thus requires appropriate access controls.
To address this last point, we now support decryption in edge workers like Cloudflare or Supabase Edge Functions. We also have browser-based encryption/decryption in the works. Both of these approaches mean data is never decrypted in the app, not even in production.