reticiulum-specification/todo.md

272 lines
15 KiB
Markdown
Raw Normal View History

# TODO
Outstanding work for the spec repo.
## Outreach
- [ ] **File a community-documentation issue on `markqvist/Reticulum`.**
Link this repo as a community-maintained byte-level spec. Ask
whether the maintainer would like to bless / link from the
official Reticulum manual. Frame it as a complement to (not a
replacement for) the existing operator-focused docs.
## Test infrastructure
- [x] **Bootstrap `test-vectors/identities.json`** — Alice + Bob
identities populated against RNS 1.2.0. Regenerator at
`tools/regen_identities.py`.
- [ ] **Bootstrap remaining test-vectors files** (`announces.json`,
`lxmf.json`, `links.json`) with the existing vectors from
`reticulum-mobile-app/reference/test-vectors.json`. Convert to
the proposed JSON format documented in `test-vectors/README.md`,
adding the regenerator scripts so future contributors can
verify vectors against newer upstream RNS releases.
- [ ] **Write the priority verifier scripts** listed in
`tools/README.md`, in this order (highest interop value first):
1. [x] `verify_destination_hash.py` — pure-function check, no RNS state needed
2. [x] `verify_packet_header.py` — bit layout + HEADER_1/HEADER_2 round-trip + originator HEADER_1→HEADER_2 conversion
3. [ ] `verify_announce_roundtrip.py` — closes the SPEC.md §4 gap (partial coverage in `verify_announce_app_data.py`)
4. [ ] `verify_token_crypto.py` — closes SPEC.md §3 gap
5. [ ] `verify_lxmf_opportunistic.py` — closes SPEC.md §5 gap
6. [ ] `verify_link_handshake.py` — closes SPEC.md §6 gap
7. [x] `verify_path_request.py` — closes SPEC.md §7.1, §7.2 gaps
8. [ ] `verify_msgpack_quirk.py` — closes SPEC.md §9.3 gap
Each verifier should remove its corresponding `⚠️ UNVERIFIED` /
`🔮 SPECULATION` callout in `SPEC.md` (per `agent.md` §1).
## Open `⚠️ UNVERIFIED` items in SPEC.md
These need either a runtime test or a stronger upstream source citation
to remove their markers:
- [x] **§2.3 Originator HEADER_1 → HEADER_2 conversion.** Verified
against RNS 1.2.0 by `tools/verify_packet_header.py`, which
seeds `Transport.path_table` with a multi-hop entry and confirms
the converted wire bytes via stubbed `Transport.transmit`.
Citation updated to `RNS/Transport.py:1074-1083`.
- [x] **§4.3 The 3-element `[name, stamp_cost, [capabilities]]`
app_data variant.** Verified against LXMF 0.9.6 by
`tools/verify_announce_app_data.py`. Finding: in this LXMF
version the producer emits a 2-element form only (the
`supported_functionality` line at `LXMF/LXMRouter.py:999` is
dead code); the parser is prepared for a 3-element form via
`compression_support_from_app_data`. SPEC.md §4.3 updated to
describe the actual current behavior.
- [x] **§7.1 path? always precedes LXMF DATA.** Verified against
LXMF 0.9.6 by `tools/verify_path_request.py`. Finding: the
preamble fires only when `not has_path()` AND method is
OPPORTUNISTIC; the retry path can fire a second `request_path`
after `MAX_PATHLESS_TRIES` (`LXMRouter.py:2571+`). SPEC.md §7.1
rewritten accordingly. Also fixed a documentation bug in §1.2
(path-request name_hash column).
- [x] **§7.4 Ratchet ring count default = 8.** False — actual upstream
default is `Destination.RATCHET_COUNT = 512` at
`RNS/Destination.py:85` in RNS 1.2.0, with
`RATCHET_INTERVAL = 30*60` (line 90) and
`RATCHET_EXPIRY = 60*60*24*30` (`RNS/Identity.py:69`).
SPEC.md §7.4 corrected.
## Open `⚠️` items needing a runtime verifier
Expand §8.3 with the full RNode air-frame split-packet protocol The previous one-sentence §8.3 was wrong about scope: it said KISS hosts treat the 1-byte header as opaque pass-through, which is misleading — the byte lives between RNodes on the LoRa air-frame, not on the KISS channel. Hosts (RNS, Sideband, etc.) never see it. Any alternative implementation that talks LoRa to an RNode must construct/parse it bit-exactly, or its TX is invisible and its RX mistakes the header for the first payload byte. New text covers: - Header byte layout: bit 7..4 random seq nibble, bit 0 FLAG_SPLIT, SEQ_UNSET=0xFF sentinel (Framing.h:105-108). - TX rules: header = random(256) & 0xF0 | (FLAG_SPLIT iff payload > 254). Both halves of a split share the same byte byte- for-byte. Split at 255 bytes total per LoRa frame; max reassembled payload 508. (RNode_Firmware.ino:716-742; Config.h:59-61.) - RX state machine: at most one buffered first-half keyed by seq nibble; four cases for inbound frames (RNode_Firmware.ino:359-446). - Reassembly timeout: upstream firmware has none (relies on subsequent traffic to evict). The clean-room repeater adds a 500ms defensive timeout (reticulum-lora-repeater/src/Radio.cpp:189-194) — implementation-private, not part of the wire spec. - Sequence-collision ceiling: 4 random bits = 1/16 collision per overlapping split-packet pair from the same sender. Don't burst. - Note that a "header rotates between transmissions" memory of this protocol is a fading recall of the per-TX random seq nibble — there is no retransmit-driven byte rotation or rechunk. LoRa TX is fire-and-forget; higher-layer retransmit just re-runs the TX path and gets a fresh random seq. todo.md gets an entry for tools/verify_rnode_split.py to lock the new §8.3 in with a runtime test. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-03 10:34:18 -04:00
- [ ] **`tools/verify_rnode_split.py` to lock in §8.3.** The RNode
air-frame split-packet protocol is now documented in SPEC.md §8.3
against direct citations in `markqvist/RNode_Firmware/Framing.h`,
`Config.h`, `Utilities.h`, and `RNode_Firmware.ino`, plus the
clean-room reimplementation in `thatSFguy/reticulum-lora-repeater/src/Radio.cpp`.
A runtime verifier would: build a 300-byte synthetic Reticulum
packet, run it through a Python implementation of the TX-side
header rules, and confirm the byte-level frames match what
`RNode_Firmware.ino:716-742` would emit (header byte high nibble
random + low-nibble FLAG_SPLIT bit, both frames sharing the same
header, split point at 255 bytes total per LoRa frame). RX-side
verifier should drive the state-table at SPEC.md §8.3 and confirm
the four reassembly cases.
- [ ] **Lock in the §6.2 / §6.3 corrections with `verify_link_handshake.py`.**
The wire-byte order of the LRPROOF body (`signature || responder_X25519_pub || signalling`,
not `link_id || responder_X25519_pub || signature || signalling`) and
the `link_id` derivation offsets (`N=2` for HEADER_1, `N=18` for HEADER_2,
not 18/34) were corrected against direct upstream source citations
(`RNS/Link.py:376`, `RNS/Packet.py:354-361`) in `SPEC.md` §6.2/§6.3
while writing `flows/send-link-lxmf.md`. They are source-cited but
not yet exercised by a runtime verifier. Add `tools/verify_link_handshake.py`
that drives an upstream LINKREQUEST → LRPROOF → ACTIVE handshake and
asserts byte-level layouts + `link_id` invariance under HEADER_1↔HEADER_2.
## Spec gaps for a functional client (priority-ordered)
The items below are missing pieces that prevent a client built only from
this spec (plus the existing flows/) from interoperating with upstream.
Tier 1 = required to talk at all to the mesh as a leaf LXMF client.
Tier 2 = required for a client that's actually useful (chat that works
in the wild). Tier 3 = required to act as a transport node / relay.
Where I've already done the source reading, I've left the file/line
citations inline so whoever picks the item up can start without
re-research.
### Tier 1 — required for a barebones leaf LXMF client to interop
Add receive-announce flow + SPEC §4.5 validation rules Closes the highest-priority Tier 1 gap. Without this, a from-scratch client can't learn any peers exist; known_destinations stays empty and every outbound message fails at recall(dest_hash). SPEC.md §4.5 (new): announce validation rules with full citations to RNS/Identity.py::validate_announce (line 496) and the dispatch path in RNS/Transport.py:1623-2024. Covers the body parse with context_flag branch, signed_data reconstruction (including the empty-bytes-not-absent ratchet rule), Ed25519 signature verification, dest_hash recomputation, public-key collision rejection, blackhole list, cache update order (known_destinations -> known_ratchets -> path_table), PATH_RESPONSE distinction, and the implementation-private SHOULD rules around ingress rate limiting, random_blob history caps, and self-announce filtering. flows/receive-announce.md: chronological walk through 9 steps from deframing to handler dispatch, with the cheap-pre-filter design (signature-checked-then-counted) called out, the burst-active ingress limiter explained against IC_BURST_FREQ_NEW=6Hz / IC_BURST_FREQ=35Hz, the path-table decision tree, and the announce_handlers fan-out with aspect_filter and PATH_RESPONSE filtering. Ends with a wire-byte diagram and a per-step source map. Two side fixes found while drafting: - SPEC.md §4.1 had random_hash described as "10 random bytes". It's actually random_hash = get_random_hash()[0:5] + int(time.time()).to_bytes(5, "big") per RNS/Destination.py:282. Transit relays parse the trailing 5 bytes via timebase_from_random_blob (RNS/Transport.py:3100) for replay-ordering decisions. - SPEC.md §2.5 contexts table was missing PATH_RESPONSE = 0x0B (RNS/Packet.py:83). flows/README.md status table updated; the priority-ordered todo list also gets a few new entries spun off from the work (send-announce, forward-announce, send-resource, path-discovery flows). Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-03 10:56:11 -04:00
- [x] **`flows/receive-announce.md` + SPEC.md §4.5 announce validation
rules.** Done. SPEC.md §4.5 covers the MUST validation rules
(body parse with `context_flag` branch, signed_data
reconstruction, signature verification, dest_hash recomputation,
public-key collision rejection, blackhole list, cache update
order, PATH_RESPONSE handling). `flows/receive-announce.md` walks
the chronology end-to-end. Side fixes: SPEC.md §4.1 corrected
(`random_hash` is 5 random bytes + 5 bytes big-endian uint40
unix_seconds, not 10 random bytes); SPEC.md §2.5 contexts table
now lists `0x0B PATH_RESPONSE`.
Add §10 Resource fragmentation + send-resource flow Closes Tier 1 #2. Without this, a client can't send any LXMF body larger than LINK_PACKET_MAX_CONTENT ≈ 360 B, can't receive a NomadNet page that doesn't fit in one MTU, and can't transfer files via rncp. SPEC.md §10 (new): full Resource fragmentation protocol with citations to RNS/Resource.py. 13 sub-sections covering preparation pipeline (metadata prefix → optional bz2 → random_hash prefix → SHA-256 over data||random_hash → link.encrypt of the WHOLE blob → part-split into SDU-sized chunks → 4-byte map_hash hashmap with collision guard within COLLISION_GUARD_SIZE = 2*WINDOW_MAX + HASHMAP_MAX_LEN), wire context inventory (RESOURCE_ADV / RESOURCE / RESOURCE_REQ / RESOURCE_HMU / RESOURCE_PRF / RESOURCE_ICL / RESOURCE_RCL), the msgpack dict for the advertisement (t/d/n/h/r/o/i/l/q/f/m), the request payload format with the hashmap_exhausted sentinel, the lazy-hashmap RESOURCE_HMU continuation that lets large hashmaps avoid breaking small-MTU links, the proof body resource_hash(32) || full_proof = SHA256(data||hash) (32) returned in a PROOF-type packet, the sliding window dynamics (WINDOW=4 → WINDOW_MAX_FAST=75 / WINDOW_MAX_VERY_SLOW=4 with rate detection), multi-segment cutover at MAX_EFFICIENT_SIZE = 1 MiB - 1 with the lazy `__prepare_next_segment` pattern, and the encryption-before-split layering that means a missing part can't be decrypted in isolation. flows/send-resource.md: 10-step chronology from RNS.Resource() construction through advertise → req/parts loop → HMU continuation → final RESOURCE_PRF → multi-segment fan-out, with a wire-byte ladder diagram and a per-step source map. Side fixes found while drafting: - SPEC.md §2.5 contexts table was wildly incomplete and had a real bug: KEEPALIVE was listed as 0xFD; upstream is 0xFA per RNS/Packet.py:87. 0xFD is actually LINKPROOF (the regular DATA-receipt context, §6.5). Replaced with the full upstream context inventory: NONE, RESOURCE_*, CACHE_REQUEST, REQUEST, RESPONSE, PATH_RESPONSE, COMMAND, COMMAND_STATUS, CHANNEL, KEEPALIVE, LINKIDENTIFY, LINKCLOSE, LINKPROOF, LRRTT, LRPROOF. - SPEC.md §6.5 reworded: "send back a PROOF packet (no context byte specifics)" → "send back a PROOF-type packet with context = LINKPROOF (0xFD)" for clarity. - The previously-numbered §10 "Test vectors" and §11 "Source map" are renumbered to §11 / §12 so the new Resource section lands in its correct protocol-stack position. agent.md §5 audit table updated accordingly. flows/README.md status table updated; receive-resource.md added as the next pending flow. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-03 11:08:40 -04:00
- [x] **SPEC.md §10 / `flows/send-resource.md`: Reticulum Resource
fragmentation.** Done. SPEC.md §10 covers the wire-level MUST
rules: 13 sub-sections from "when Resource runs" through wire
contexts (ADV / REQ / RESOURCE / HMU / PRF / ICL / RCL),
hashmap collision-guard, sliding window, multi-segment cutover
at MAX_EFFICIENT_SIZE = 1 MiB - 1, and the encryption-then-split
layering. `flows/send-resource.md` walks the chronology in 10
steps with a wire-byte ladder diagram. Side fixes during the
drafting: SPEC.md §2.5 contexts table now lists ALL upstream
contexts (was missing all RESOURCE_*, REQUEST/RESPONSE,
COMMAND, CHANNEL, LINKIDENTIFY, LINKCLOSE, LRRTT entries) and
corrects KEEPALIVE from 0xFD (which is actually LINKPROOF) to
0xFA per `RNS/Packet.py:87`. SPEC.md §6.5 wording updated to
use the correct LINKPROOF context name. The previously-existing
§10 "Test vectors" and §11 "Source map" were renumbered to §11
and §12 to put §10 in the protocol-stack flow.
- [ ] **SPEC.md §6.5 expansion: regular (non-LRPROOF) PROOF body.** The
mandatory PROOF receipt for every CTX_NONE Link DATA packet. Body
is `packet_hash(32) || signature(64)` (`RNS/Link.py::prove_packet`
line 384-393), with a hardcoded explicit-mode comment hinting at
a future implicit-mode toggle that elides the packet_hash prefix.
Adding a `tools/verify_proof_packet.py` that runs a real link
transfer and asserts the proof body shape is the right
verification.
- [ ] **SPEC.md §6 sub-section: 3-byte MTU/mode signalling field.**
Present on LINKREQUEST and LRPROOF iff
`Reticulum.link_mtu_discovery() == True` and the next-hop
interface advertises an HW MTU. Encode/decode helpers at
`RNS/Link.py::signalling_bytes` line 148; consumers at
`mtu_from_lr_packet` / `mode_from_lr_packet` /
`mtu_from_lp_packet` / `mode_from_lp_packet`. Spec currently
shows this slot as "[signalling(3)]" with no byte definition —
a client that emits a wrong format gets wrong MTU on the link.
- [ ] **SPEC.md §7.2 expansion + new flow `flows/path-discovery.md`:
path-response announce vs periodic announce.** When a node
fulfills a `path?` request it emits an announce with
`path_response=True`, which sets `context = PATH_RESPONSE = 0x0B`
on the announce packet (`RNS/Packet.py:83`). Receivers
distinguish via `packet.context == RNS.Packet.PATH_RESPONSE`
(`RNS/Transport.py:1989-1991`); announce handlers default to
ignoring path-responses unless they set
`receive_path_responses = True` on themselves. Spec mentions §7.2
"respond by re-announcing" but doesn't name the wire context.
- [ ] **SPEC.md §1.3 expansion: identity on-disk format.** §1.3 names
the byte order (Ed25519 first, X25519 second, opposite of the
public-key concat) but not the file structure. `RNS/Identity.py::to_file`
is the reference. Without this, identities can't be exported /
imported across implementations.
### Tier 2 — required for a client to be useful in the wild
- [ ] **SPEC.md: Propagation node protocol.** Offline message retrieval
via store-and-forward propagation nodes. Without this, every
message requires both peers online simultaneously. Authoritative
source: `LXMF/LXMRouter.py::process_propagated`, the
`lxmf.propagation` peering exchange (`peer()` / `sync()` between
nodes — `LXMRouter.py:1892+, 2118+`). The `propagated` method is
already in `LXMessage.py` but the wire protocol between
propagation nodes is undocumented. Cross-flow:
`flows/send-propagated-lxmf.md` (already a `⏳` entry in
`flows/README.md`).
- [ ] **SPEC.md §6 expansion: KEEPALIVE / link teardown protocol.**
`CTX_KEEPALIVE = 0xfd` packets — exact wire body, exact cadence
(`Link.KEEPALIVE` constant), exact teardown packet (`Link.PROOF`
context). Real clients drop links incorrectly without this.
- [ ] **SPEC.md §5.x (new): LXMF stamps + tickets for spam control.**
`LXMF.Stamp` (proof-of-work field in the optional 5th element of
the msgpack payload), `FIELD_TICKET` lookup. Modern Sideband 1.x
treats missing-stamp messages as spam in the UI. Spec currently
doesn't mention stamps at all. Authoritative source:
`LXMF/LXMessage.py::validate_stamp`, `LXMF/LXMRouter.py:1741-1774`
(the stamp-check branch in `lxmf_delivery`).
- [ ] **SPEC.md §13 (new): NomadNet page protocol.** Distinct from
LXMF — pages fetched over a Link with `context = CTX_REQUEST (0x09)`
/ `CTX_RESPONSE (0x0a)` (already in §2.5 contexts table). Request
body is a path string + field map; response is a body bytes blob.
Without this, a client can do LXMF chat but can't render NomadNet
content (nodes serving content, telemetry, micron pages).
- [ ] **SPEC.md §1.4 (new): GROUP destinations.** `RNS.Destination.GROUP`
type uses symmetric AES-256-CBC with a pre-shared key; different
encrypt/decrypt paths in `RNS/Destination.py:601+` (`prv` is a
symmetric-key wrapper, not an X25519 priv). Almost no clients
implement this but the protocol allows it.
- [ ] **SPEC.md §8.4 (new): CSMA / airtime tracking.** LoRa-only —
carrier-sense + random backoff that prevents transmitter
collisions on shared channel. The clean-room repeater explicitly
flags "no CSMA" as a phase-2 simplification. A serious LoRa
client needs `RNS.Reticulum.ANNOUNCE_CAP`-aware backoff and the
`airtime_bins` accounting from `RNode_Firmware.ino:683-712`.
- [ ] **SPEC.md §8.5 (new): RNode KISS configuration handshake.**
Beyond §8.3 (split-packet protocol), a client opening an RNode
drives `CMD_DETECT` / `CMD_FREQUENCY` / `CMD_BANDWIDTH` /
`CMD_SF` / `CMD_CR` / `CMD_TXPOWER` / `CMD_RADIO_STATE` over KISS
to bring up the radio. All defined in `RNode_Firmware/Framing.h:24-95`.
Spec just says "send Reticulum packets via CMD_DATA" — that's
not enough.
- [ ] **SPEC.md §6.5 second sub-bullet: implicit vs explicit proof
mode.** `RNS.Reticulum.should_use_implicit_proof()` mode trims
the proof body to just the signature (no `packet_hash` prefix),
saving 32 bytes. `RNS/Link.py:386-389` has the explicit form
hard-coded with the implicit branch commented out, but at least
one upstream branch toggles it — a client that hard-codes the
explicit form will eventually meet a peer in implicit mode.
### Tier 3 — required to act as a transport node / relay
- [ ] **SPEC.md §7.7 (new): DATA forwarding rules.** Forwarding non-
local DATA per `path_table[dest][NEXT_HOP]`, with hop increment,
MTU-fit check, blackhole avoidance, and IFAC re-signing.
Currently mentioned only obliquely in §2.3 / §7.6. The full
forwarding logic is the bulk of `RNS/Transport.py::inbound`'s
~800-line dispatch table at lines 1499-1620. The repeater repo
patches microReticulum to enable this — see commit
`Add DATA and PROOF forwarding patches for transport repeating`.
- [ ] **SPEC.md §4.6 (new): ANNOUNCE rebroadcasting.** Including the
announce-cap (`RNS.Reticulum.ANNOUNCE_CAP`, default 2% airtime),
the announce queue, the `path_responses` cache, and the
`random_blob` history that lets a relay drop replays. Most of
`RNS/Transport.py:1196-1300, 1810-1969`.
- [ ] **SPEC.md §7.8 (new): path table management.** TTL-based expiry
(`Transport.AP_PATH_TIME`, `ROAMING_PATH_TIME`, `DESTINATION_TIMEOUT`),
eviction on stale-link, persistence-across-reboot file format.
Hooks: `RNS/Transport.py:747-769` (stale_paths accumulator) and
the `paths` file under `storagepath`.
- [ ] **SPEC.md §7.9 (new): tunnels and shared-instance protocol.**
`tunnels`, `discovery_path_requests`, `RNS/Reticulum.py::is_connected_to_shared_instance`
— how a process talks to a co-resident `rnsd`. Spec's §7.6
covers one symptom (TCP OUT default) but not the actual
shared-instance wire format.
- [ ] **SPEC.md §6.x (new): reverse table + link transport.** When a
Link's path crosses a relay, the relay must forward both
directions of every Link DATA + PROOF using
`Transport.reverse_table` (`RNS/Transport.py:2087-2204`).
Distinct from path-table forwarding — different lookup, different
lifecycle.
## Spec polishing (lower priority)
- [ ] **Split `SPEC.md` into per-layer files** as the document grows
past ~1500 lines. Suggested layout per `README.md`:
`00-overview.md`, `01-packet-header.md`, `02-identity.md`,
`03-announce.md`, `04-token-crypto.md`, `05-lxmf.md`,
`06-link.md`, `07-resource.md`, `08-transport.md`,
`09-paths-and-discovery.md`, `10-implementation-gotchas.md`.
- [ ] **Add a "last-verified-against-rns" line** to SPEC.md
frontmatter (per `agent.md` §7) so readers know which RNS
version the spec was tested against.