path: root/markup/api.md

# System Transparency Log
This document provides a sketch of System Transparency (ST) logging.  The basic
idea is to insert hashes of system artifacts into a public, append-only, and
tamper-evident transparency log, such that any enforcing client can be sure that
they see the same system artifacts as everyone else.  A system artifact could
be an operating system image, a Debian package, or generally just a checksum of
something opaque.

An ST log can be implemented on-top of
[Trillian](https://trillian.transparency.dev) using a custom STFE personality.
For reference you may look at Certificate Transparency (CT) logging and
[CTFE](https://github.com/google/certificate-transparency-go/tree/master/trillian/ctfe),
which implements [RFC 6962](https://tools.ietf.org/html/rfc6962).

We take inspiration from from RFC 6962 and its follow-up specification [RFC
6962/bis](https://datatracker.ietf.org/doc/draft-ietf-trans-rfc6962-bis/).

## Log parameters
A log is defined by the following immutable parameters:
- Log identifier: `SHA256(public key)`, see RFC 6962
[§3.2](https://tools.ietf.org/html/rfc6962#section-3.2). TODO: use KID instead.
- Public key: DER encoding of the key represented as `SubjectPublicKeyInfo`
- Supported signature schemes: a list of signature schemes that the
log recognizes.  Possible values are defined in RFC 8446,
[§4.2.3](https://tools.ietf.org/html/rfc8446#section-4.2.3).  Submitters must
use a signature algorithm that the log supports.
- Signature scheme: the signature scheme that the log uses to sign tree heads
and debug info statements.
- Base URL: where can this log be reached?  E.g., example.com:1234/log

Note that **there is no MMD**.  The idea is to merge added entries as soon as
possible, and no client should trust that something is logged until an inclusion
proof can be provided that references a trustworthy STH. 

Moreover, we use the same hash strategy as described in RFC 6962: SHA256 with
`0x00` as leaf node prefix and `0x01` as interior node prefix.

## Minimum acceptance criteria
A log should accept a submission if it is:
- Well-formed, see below.
- Digitally signed
	- Proves who submitted an entry for logging
	- Verification key must be registered in the log as a namespace

## Data structure definitions
We encode everything that is digitally signed as in [RFC
5246](https://tools.ietf.org/html/rfc5246).  Therefore, we use the same
description language for our data structures.  A definition of the log's Merkle
tree can be found in RFC 6962, see
[§2](https://tools.ietf.org/html/rfc6962#section-2).

### Repurposing `TransItem` as `StItem`
A general-purpose `TransItem` is defined by RFC 6962/bis.  Below we define our
own `TransItem`, but name it `STItem` to emphasize that they are not the same.
Some definitions are re-used and others are added.

```
enum {
	reserved(0),
	signed_tree_head_v1(1), // defined in RFC 6962/bis, §4.10
	signed_debug_info_v1(2), // defined below, think "almost SCT"
	consistency_proof_v1(3), // defined in RFC 6962/bis, §4.11
	inclusion_proof_v1(4), // defined in RFC 6962/bis, §4.12
	checksum_v1(5), // defined below, think "leaf data"
	(65535)
} StFormat;

struct {
	StFormat format;
	select (format) {
		case signed_tree_head_v1: SignedTreeHeadV1;
		case signed_debug_info_v1: SignedDebugInfoV1;
		case consistency_proof_v1: ConsistencyProofV1;
		case inclusion_proof_v1: InclusionProofV1;
		case checksum_v1: ChecksumV1;
	} message;
} StItem;
```

### Namespace
The submitter's verification key is used to establish a _namespace_.  Added
log entries must be signed by a registered namespace, such that anyone that
observes the log can determine which artifact hashes belong to which namespaces.
```
enum {
	reserved(0),
	ed25519_v1(1),
	(65535)
} NamespaceFormat;

struct {
	NamespaceFormat format;
	select (format) {
		case ed25519_v1: Ed25519V1;
	} message;
} Namespace;
```

Credit: inspired by Keybase's [KID format](https://keybase.io/docs/api/1.0/kid).

#### Ed25519V1
At this time the only supported key type is Ed25519 as defined by [RFC
8032](https://tools.ietf.org/html/rfc8032).  The namespace field contains the
full verification key.
```
struct {
	opaque namespace<32>; // public key
} Ed25519V1;
```

#### Other
In the future we will support other key types, such as RSA.  For example, we
could add [RSASSA-PKCS1-v1_5](https://tools.ietf.org/html/rfc3447#section-8.2)
as follows:
1. Add `rsa_v1` format and RSAV1 namespace.  This is what we would register on
the server-side such that the server knows the namespace and complete key.
```
struct {
	opaque namespace<32>; // key fingerprint
	// + some encoding of public key
} RSAV1;
```
2. Add `rsassa_pkcs1_5_v1` format and `RSASSAPKCS1_5_v1`.  This is what the
submitter would use to communicate namespace and RSA signature mode.
```
struct {
	opaque namespace<32>; // key fingerprint
	// + necessary parameters, e.g., SHA256 as hash function
} RSASSAPKCS1_5V1;
```

Another option is to just never bother with key fingerprint, i.e., use the
complete (encoded) RSA key as the namespace.  Makes the leaf a lot larger
though.

### Merkle tree leaf types
In the future there might be several types of leaves.  Say, one for operating
system packages, another one for Debian packages, and a third one for
general-purpose checksums.  For now we only define the latter.

TODO: scope of this spec should only be checksum

#### Checksum

```
struct {
	opaque package<1..2^8-1>; // package identifier
	opaque checksum<1..64>; // hash of some artifact
	Namespace namespace;
} ChecksumV1;
```

A checksum entry contains a package identifier such as `foobar-1.2.3` and an
artifact hash.  For example, the checksum type could be used by Firefox to
[enforce public binary logging before accepting a new software
update](https://wiki.mozilla.org/Security/Binary_Transparency).  It is assumed
that the entities relying on the checksum type know how to find the artifact
source (if not already at hand) and then reproduce the logged hash from it.

Namespace is used to determine who this artifact hash belongs to.  Note that we
do not connect namespaces to real-world identities.  It is just _a namespace_.

### Signed Debug Info
RFC 6962 uses Signed Certificate Timestamps (SCTs) as promises of public
logging within a time known as the Maximum Merge Delay (MMD).  We provide no
such promise: a Signed Debug Info (SDI) is an intent to log because the
submitter is authorized to do so and the entry appears to be valid.  It will be
merged into the log's Merkle tree as soon as possible on a best-effort basis.
If an unexpected delay is encountered, the submitter can present the issued SDI
to the log operator (who can then investigate the underlying reason further).
```
struct {
	LogID log_id; // defined in RFC 6962, basically SHA256(pub key)
	opaque message<1..2^16-1> // debug string that is only meant for the log
	opaque signature <1..2^16-1; // computed over a leaf-type StItem
} SignedDebugInfoV1;
```

The signature's encoding follows from the log's signature algorithm parameter,
e.g., `ed25519(0x0807)` refers to [RFC
8032](https://tools.ietf.org/html/rfc8032).  A complete list of signature
schemes and their interpretations can be found in RFC 8446,
[§4.2.3](https://tools.ietf.org/html/rfc8446#section-4.2.3).

TODO: when log id is namespace this information is already communicated.
TODO: remove SDI?

## Public endpoints
Clients talk to the log with HTTPS GET/POST requests.  POST parameters
are JSON objects, GET parameters are URL encoded, and serialized data is
expressed as base-64.  See details in as in RFC 6962,
[§4](https://tools.ietf.org/html/rfc6962#section-4).

TODO: remove json
TODO: and b64?

Unless specified otherwise, the data in question is serialized.

### add-entry
```
POST https://<base url>/st/v1/add-entry
```

Input:
- item: an `StItem` that corresponds to a valid leaf type.  Only
`checksum_v1` at this time.
- signature: covers the submitted item.

Output:
- an `StItem` structure of type `signed_debug_info_v1` that covers the added
item.

### get-entries
```
GET https://<base url>/st/v1/get-entries
```

Input:
- start: 0-based index of first entry to retrieve in decimal.
- end: 0-based index of last entry to retrieve in decimal.

Output:
- an array of objects, each consisting of
	- leaf: `StItem` that corresponds to the leaf's type.
	- signature: signature that covers the retrieved item using the below
	signature scheme.

### get-namespaces
```
GET https://<base url>/st/v1/get-namespaces
```

No input.

Output:
- an array of base-64 encoded namespaces that the log accept. TODO: format?

### get-proof-by-hash
```
GET https://<base url>/st/v1/get-proof-by-hash
```

Input:
- hash: a base-64 encoded leaf hash.
- tree_size: the tree size that the proof should be based on in decimal.

The leaf hash value is computed as in RFC 6962/bis,
[§4.7](https://datatracker.ietf.org/doc/html/draft-ietf-trans-rfc6962-bis-34#section-4.7).

Output:
- an `StItem` of type `inclusion_proof_v1`.  Note that this structure includes
both the leaf index and an audit path for the tree size.

### get-consistency-proof
```
GET https://<base url>/st/v1/get-consistency-proof
```

Input:
- first: the `tree_size` of the older tree in decimal.
- second: the `tree_size` of the newer tree in decimal.

Output:
- an `StItem` of type `consistency_proof_v1` that corresponds to
the requested tree sizes.

### get-sth
```
GET https://<base url>/st/v1/get-sth
```

No input.

Output:
- an `StItem` of type `signed_tree_head_v1`, which corresponds to the most
recently known STH.