The new truncation test's doc landed between backend_tool_call_position_
stable's doc and its #[test] attribute (clippy: duplicated attribute).
Each test owns its own doc block again.
Cross-provider audit R4+M3: the Responses leg passed tool-call ids
verbatim (call_id on both function_call and function_call_output) while
the Messages leg sanitized — and its closure used Unicode
is_alphanumeric, letting CJK ids through to Anthropic's ASCII-only
contract, with no empty-id fallback. Both providers enforce
[A-Za-z0-9_-]+ (the codex 400's own words). One module-scope
sanitize_tool_call_id now serves both builders, ASCII-only, empty → "_",
applied identically on call+result so pairing survives.
Part 4 of the cross-provider replay audit.
Verified: 6081 tests green across the four crates, clippy clean.
Cross-provider audit R3: reasoning captured on stateful api.openai.com
sessions (no include requested) carries a server-issued rs_* id but NO
encrypted_content; replaying that bare reference to the stateless
(store:false) codex backend points at server state chatgpt.com does not
have. adapt_body_for_codex_backend step 3 drops such items (encrypted
ones pass through verbatim). Capture-side include for the API-key path
is deferred: the typed CreateResponse is shared with the xai Responses
leg and changing its bytes needs separate validation.
Part 3 of the cross-provider replay audit.
Verified: sampling-types+sampler+chat-state+shell all green.
The multibyte-truncation test used 80 chars — below the 100-char cap, so
the truncation assertion failed (the previous commit's suite count was
misread; the FIX itself was correct and the panic repro held). 120 chars
now exercises both the boundary safety and the truncation.
The code-interpreter preview truncated at BYTE 100 — a guaranteed panic
on any CJK/emoji boundary in interpreted code. One poisoned history item
then crashed every subsequent request build on every backend (the
summary feeds both the Messages and ChatCompletions builders). Truncate
at 100 chars via char_indices instead. Panic pinned by test.
Part 2 of the cross-provider replay audit.
Verified: sampling-types 287 tests green.
Root cause of 400 Invalid 'input[N].id': '' on chatgpt.com/backend-api/
codex/responses: the Responses input builder replayed every stored
Reasoning item verbatim, and rs::ReasoningItem.id serializes
unconditionally — so foreign items (Messages-captured Anthropic
signatures, chat-completions-synthesized reasoning, stream-delta
fallbacks, legacy upgrades — all id '') reached the wire with an empty
id the API rejects. This also self-poisoned pure codex sessions whose
reasoning arrived only as deltas.
A native item always carries a server-issued rs_* id: empty id = foreign
= unusable by any Responses provider = dropped at the builder — the
exact mirror of the Messages builder's prune_replayed_thinking. The
encrypted-only fixture that pinned the poison shape now uses a native id
(the pass-through case it always meant to cover).
Part 1 of the cross-provider replay audit (Pi transform-messages
policy: builders emit only items valid for their target).
Verified: sampling-types 286 + sampler/chat-state/shell 5791 green.
Root cause of 400 {"detail":"System messages are not allowed"} at
chatgpt.com/backend-api/codex/responses (both platforms): the codex
adaptation covered only IDENTITY HEADERS (originator/OpenAI-Beta/UA/
chatgpt-account-id) — the BODY still carried the system prompt as
role:system input items, which the codex backend rejects outright. Its
system channel is the top-level field, and stateless
(store:false) reasoning replay requires
include:["reasoning.encrypted_content"] — both per the same reference
the headers were ported from (official Codex CLI + Pi's
api/openai-codex-responses.ts).
New adapt_body_for_codex_backend (kigi-sampling-types): hoists every
system input item into (order preserved, appended to any
existing instructions; string and parts content shapes) and requests
encrypted reasoning. Idempotent. Applied at both Responses send paths,
openai_codex-GATED — the API-key path stays byte-identical
(pinned by a control wire test).
Tests: adapter unit tests (hoist+include, no-system no-op, idempotence)
plus two mock-server wire tests (codex body has no system role,
instructions + include present; plain Responses body unchanged).
Verified: sampling-types + sampler + chat-state + shell 6076 tests
green, clippy clean.
Root cause of 'Invalid signature in thinking block' (400 at
messages.1.content.0, Claude models): build_messages_request replayed
EVERY stored Reasoning item as a thinking block with no origin check —
history synthesized by other backends (encrypted_content: None → the
mandatory signature field serialized as ""), Responses-API tco_* blobs
(signature bytes, no text), and blocks signed by a DIFFERENT model after
a mid-session /model switch. Anthropic validates every replayed
signature (model-bound), so such histories 400 deterministically. The
platform split was circumstantial: Windows sessions started on the
default model and switched to Claude; macOS sessions were Claude-native
from turn 1. Adversarially verified — no platform-divergent byte path
exists in capture, storage, or replay.
New prune_replayed_thinking pass (Pi/Claude Code replay policy): keep
exactly the final assistant message's thinking when its tool loop is
still open (request ends on the tool results — an open loop can never
span a model switch) and the block is genuinely signed; strip every
other thinking block (the API ignores valid prior-turn thinking and
rejects invalid). Assistant messages emptied by the strip (thinking-only
aborted turns) are removed — empty content arrays are rejected too.
Tests: three unit tests pin strip-outside-loop (unsigned, tco_*, stale
signed), keep-in-active-loop (verbatim text+signature at content.0), and
emptied-message removal; the legacy-upgrade integration test now proves
both wire fidelity in the active loop AND stripping once the loop
closes.
Verified: sampling-types + sampler + chat-state + shell 6072 tests
green, clippy clean.
29th platform `openai-codex` (uses_oauth, Responses wire). PKCE-localhost login at
auth.openai.com (client app_EMoamEEZ73f0CkXaXp7hrann, redirect localhost:1455/auth/
callback, form token exchange, fresh-random state) reusing the claude-pro-max flow;
OAuthFlow::PkceLocalhost gained a redirect_path and OAuthConfig an authorize_extra
(empty elsewhere, so claude/xai/copilot authorize URLs stay byte-identical).
Codex-specific: the access token is a JWT carrying chatgpt_account_id, which becomes
the `chatgpt-account-id` inference header. It is derived STATELESSLY from whichever
bearer rides each request (so a rotated token needs no persisted field), and BOTH
login and refresh fail fast when the claim is absent — gated on the explicit
OAuthConfig.requires_chatgpt_account_id fact, never inferred from the token-body
encoding (a plain form endpoint is the OAuth norm and must not inherit this).
Inference rides the existing Responses wire at chatgpt.com/backend-api/codex →
/responses, with codex headers (chatgpt-account-id, originator, OpenAI-Beta
responses=experimental, codex UA) gated on SamplerConfig.openai_codex so API-key
`openai` stays byte-identical; store:false was already the global Responses default.
Catalog is HARDCODED (no live endpoint exists for this backend; read from the
official Codex CLI's model cache): gpt-5.6-sol/terra/luna + gpt-5.5, ctx 272000,
each with its real reasoning levels (low..ultra — ReasoningEffort gained Ultra).
Excluded: gpt-5.3-codex-spark (supported_in_api=false), gpt-5.4/-mini and
codex-auto-review (hidden) — they would list but fail at inference. The fetch
short-circuits before any HTTP; Kigi never shells out to the codex CLI or reads
~/.codex.
Security review fixes: redact any `account-id` header from request logs (it was
reaching debug logs), strict 3-segment JWT check (fail closed), refresh no longer
fails open on a missing claim. Inherits leak-safe pooled routing (scope
oauth/openai-codex) — never the Kimi token. Full gate green (234 suites, 0 warnings).
28th platform `github-copilot` (uses_oauth, ChatCompletions wire). Two-stage auth:
RFC-8628 GitHub device flow (client Iv1.b507a08c87ecfe98, scope read:user, errors
in a 200 body) mints the DURABLE github token; a GET api.github.com/copilot_internal/
v2/token exchange re-mints the SHORT-LIVED copilot token. Persisted as key=copilot
token, refresh_token=github token, expires_at=copilot expiry; the "refresh" is a
copilot-token re-mint (not a refresh_token grant), dispatched via
OAuthTokenBody::GithubCopilotExchange in the generic refresher.
VS Code editor-identity headers on /models + /chat/completions, gated on
SamplerConfig.github_copilot / PlatformId::sends_copilot_editor_headers() so every
other ChatCompletions provider stays byte-identical. Live /models filtered
(parse_github_copilot_listing) to the openai-completions-served models: keep iff
model_picker_enabled && policy.state!="disabled" && tool_calls!=false AND not a
claude-4.x/5.x (messages) or gpt-5/oswe/mai- (responses-only) id — those need
per-model wire routing (documented debt), excluded rather than mis-routed.
Inherits the leak-safe pooled routing (scope oauth/github-copilot); its bearer/
refresh/api_key never touch the Kimi token (regression test added). Fail-fast on
an out-of-range copilot expires_at (would otherwise silently 401 mid-session).
Adversarial security review: GO, no CRITICAL/HIGH. Known limitation: Pi's
per-model policy-enablement POST is not ported (documented in AGENTS.md).
27th registry variant, 2nd subscription-OAuth provider. Log in with a Claude
Pro/Max subscription via PKCE authorization-code + S256 (loopback callback on
127.0.0.1:53692, with a manual code-paste fallback), then use it against
api.anthropic.com — reusing the existing Anthropic Messages wire + Anthropic
listing + the multi-provider OAuth foundation (dbce6bf). Sourced from Pi
(earendil-works/pi auth/oauth/anthropic.ts): client 9d1c250a..., authorize
claude.ai/oauth/authorize, token platform.claude.com/v1/oauth/token, scope
'…user:inference user:sessions:claude_code…'.
New machinery (foundation handles token routing — claude-pro-max is a
uses_oauth platform so its bearer/refresh/api_key already route to its own
pooled manager, never Kimi):
- OAuthConfig gains flow{DeviceCode|PkceLocalhost} + token_host + token_body
{Form|JSON}; xai/kimi rows unchanged (DeviceCode/Form).
- auth/oauth_pkce.rs: PKCE S256 wire — loopback listener with STRICT state
validation (CSRF, fail-closed), manual-paste fallback, JSON code→token
exchange + rotating-refresh. Never logs code/verifier/tokens.
- Messages OAuth adaptation gated on SamplerConfig.anthropic_oauth (true only
for a claude-pro-max managed key): Authorization: Bearer + anthropic-beta
oauth + user-agent claude-cli + x-app cli, and the required 'You are Claude
Code' system prefix. API-key anthropic/minimax Messages requests are
BYTE-IDENTICAL (regression-guarded).
- Live /models under the OAuth Bearer + oauth-beta headers (Anthropic listing,
enriched from models.dev anthropic); persistent 401 → 0 models + WARN, NO
hardcoded fallback list (honest failure).
Adversarial review: no blocking findings (secret handling, CSRF/state, the
anthropic_oauth gate, token routing, non-regression all CONFIRMED). Full gate
green. Registry at 27; picker updated. Residual (unverifiable without a real
Claude Pro/Max account): whether GET /v1/models accepts the OAuth bearer, and
the real endpoint's acceptance of the OAuth Messages request.
The 13th registry row: id "cerebras", CEREBRAS_API_KEY > auth.json
"cerebras" scope, https://api.cerebras.ai/v1 with KIGI_CEREBRAS_BASE_URL
override, Bearer, ChatCompletions, enrichment-backed metadata
(models_dev_id cerebras).
Cerebras' catalog is all chat LLMs (no embedding/tts pollution) and its
/models is minimal (ids only), so restrict_to_enriched=FALSE: keep every
live model, enrich the known ones (context + effort menus low/medium/high),
unknown ones keep the default context. The e2e pins this enrich-without-
restrict path (new — prior enrichment providers all used restrict=true).
Review caught a likely-DOA defect: Cerebras uses strict
additionalProperties:false validation (confirmed 400-rejecting store,
maxTokens, thinking, nested reasoning_content), and stream_options is not
in its schema — so Passthrough (which keeps the stream_options.include_usage
kigi injects on every streaming request) would very likely 400 all
streaming. Generalized ChatCompat::Mistral -> ChatCompat::StrictOpenAi
(serde alias "mistral" keeps pre-rename persisted sessions loading), which
strips stream_options + private fields for any strict OpenAI-compat
validator; both Mistral and Cerebras now map to it. Future strict-validator
candidates (NVIDIA/Azure/Xiaomi/OpenCode) noted for the same check.
reasoning_effort (incl. "none") passes through; /v1/models requires auth
so key validation works; console cloud.cerebras.ai.
The 8th registry row: id "mistral", MISTRAL_API_KEY > auth.json
"mistral" scope, https://api.mistral.ai/v1 with KIGI_MISTRAL_BASE_URL
override, enrichment-backed metadata with the tool-calling listing
restriction (embed/moderation/OCR noise).
Mistral is NOT a pure-pattern provider — an adversarial review found two
doc-confirmed blockers that no test exercises (no e2e covers a chat POST),
so the gate-green registry row alone would have shipped it DOA. A research
workflow pinned the exact wire shapes against the mistralai/client-python
SDK source (adversarially verified), then both were fixed:
1. stream_options 422: Mistral's strict Pydantic validator rejects the
stream_options.include_usage field kigi injects on every streaming
request (the SDK's request model has no such field). New
ChatCompat::Mistral dialect strips it (plus the kigi-private message
fields, like Passthrough). Streaming usage falls back to token
estimation.
2. Reasoning content arrays: Mistral reasoning models return content as
Union[str, List[ContentChunk]] on both streaming and non-streaming,
which the flat Option<String> path could not decode -> aborted turn.
A UNIVERSAL lenient deserializer (#[serde(from = "Raw..")] on
ChatResponseMessage + ChatChunkDelta) accepts string-or-array, routing
{type:text} chunks to the answer and the nested text of {type:thinking}
chunks to reasoning_content, tolerant of the OPEN chunk union (unknown
types ignored, never fatal). String content stays byte-identical for
every other provider (kimi/deepseek/groq/BYOK).
Review refuted all seven attack lines (no regression, no crash, exhaustive)
and flagged one coverage gap, now closed: a stream-consumer integration
test drives a full thinking -> transition -> answer chunk sequence and
proves it yields the same reasoning-sibling + assistant-answer result as
the reasoning_content string path.
Also folds a verified quirk matrix for all 23 remaining API providers into
providers-plan.md, tiered by real difficulty (self-enriching OpenRouter/
Vercel; bare-array Together listing; Messages-dialect MiniMax reusing the
Anthropic machinery; non-Bearer Azure/Bedrock; router wildcards; the OAuth
block).
The 6th registry row: id "deepseek", DEEPSEEK_API_KEY > auth.json
"deepseek" scope, base https://api.deepseek.com (chat rides
{base}/chat/completions per official docs) with KIGI_DEEPSEEK_BASE_URL
override, enrichment-backed metadata (1M context, 384k output cap,
high/max effort menu).
Structural fix the cycle exposed: kigi's Kimi-specific body adaptation
ran UNCONDITIONALLY on every ChatCompletions request. New ChatCompat
dialect, declared per platform row and threaded through SamplerConfig,
ClientDefaults, and the session-persisted SamplingConfig (serde-default
Kimi keeps restored pre-field sessions and BYOK endpoints byte-identical;
production persist seams copy it; subagents inherit it):
- Kimi: full legacy pipeline (dispatch ≡ legacy pinned)
- DeepSeek: thinking:{type, reasoning_effort} per api-docs.deepseek.com
(server maps low/medium→high, xhigh→max itself; none disables; absent
leaves the server default)
- Passthrough: OpenAI-style reasoning_effort scalar untouched (unblocks
Groq and the rest of the OpenAI-compatible list)
Review-confirmed release blocker fixed: kigi replays Kimi's
reasoning_content (and its private model_id) on input assistant
messages — Kimi consumes these, but DeepSeek documents input
reasoning_content as prefix-mode-only (historically a 400) and other
providers don't know either field. The DeepSeek and Passthrough arms now
strip both; Kimi's own pipeline is untouched. Pinned on both message
shapes.
The 5th registry row: id "anthropic", ANTHROPIC_API_KEY > auth.json
"anthropic" scope, api.anthropic.com/v1 with KIGI_ANTHROPIC_BASE_URL
override, Messages dialect. Two new spec dimensions most future rows
reuse: ListingDialect (Anthropic's /v1/models wants x-api-key +
anthropic-version headers, ?limit=1000, and its own response shape) and
PlatformKeyHeader (Bearer vs x-api-key across listing/validation/
inference, with auth_scheme stamped onto entries).
The 2026 Anthropic listing serves real metadata: the adapter maps
max_input_tokens, per-level effort capabilities (low..max as the menu,
xhigh/max distinct), thinking/image flags — and enrichment fills only
genuine wire gaps (e2e pins wire-1M beating enrichment, and a zero
context filled to 200k).
Two review-confirmed defects fixed red-green:
- Output caps were dropped at three layers, so every sub-128K-output
model (64k Haiku, legacy models) would 400 on EVERY request against
the sampler's 128K max_tokens default. Wire max_tokens and enrichment
limit.output now flow to entry.max_completion_tokens.
- An explicit wire effort-decline was indistinguishable from wire
silence, letting enrichment inject effort menus pre-4.6 models reject
(adaptive thinking 400). The adapter now emits a decline sentinel
(support:false) that enrichment respects — proven end to end.
Also: the Messages client now sends anthropic-version (previously never
sent — real api.anthropic.com rejects such requests; pinned across all
three scheme/backend quadrants), key validation builds per-key-header
requests, missing listing data fails fast, empty-id ghosts drop with a
warning, kimi-import recognizes api.anthropic.com as built-in
automatically.
The 4th registry row: id "openai", OPENAI_API_KEY env > auth.json
"openai" scope (login picker/paste/validation all registry-generic —
zero TUI changes needed, pinned by the picker test), base
https://api.openai.com/v1 with KIGI_OPENAI_BASE_URL override, Responses
dialect via the new PlatformWireApi spec field, enrichment-backed
metadata (wire_serves_metadata=false).
OpenAI's GET /v1/models returns bare ids and is polluted with
tts/whisper/embeddings entries: the listing is restricted to
enrichment-known TOOL-CALLING models (review caught that membership
alone admitted models.dev-known embeddings models, which would 400 on
every agentic request; dropped ids are debug-logged for launch-day
diagnosability). Context windows, effort menus, display names, and
thinking capability come from the enrichment pipeline — wiremock e2e
pins the full contract: polluted live listing + models.dev →
one Responses-backed chat model with a 400k documented context window.
Responses max-effort wiring (closes the P0c-1 debt): canonical effort
rides a CreateResponseWrapper sidecar and patch_reasoning_effort writes
it onto the serialized body at both send sites (all seven levels pinned,
xhigh/max distinct, summary preserved); normalize_effort_echo drops
echoes async-openai's typed enum cannot represent at both the non-stream
and SSE parse seams; the dead typed to_responses_api converter is
deleted. Kimi/moonshot stay byte-identical (ChatCompletions untouched,
wire_api maps to the same default; kimi wire tests green).
kimi-import now recognizes ANY registry platform host as built-in
(was hardcoded moonshot), covering openai and future rows.
OpenAI (Responses) and Anthropic (Messages) treat xhigh and max as
DISTINCT effort levels in 2026, and the Kimi K3 wire's top tier is max —
the old parse alias (max→Xhigh) conflated them. Canonical Max now exists:
parse/as_str/serde split, Messages mapping sends xhigh and max as their
own tokens (was Xhigh→"max"), and the K3 menu token max carries
canonical Max end to end.
Kimi wire is byte-identical in all four flows (menu pick, restored
legacy xhigh session, --reasoning-effort flag, /effort command) —
adversarially traced and pinned: kimi_compat's string-level xhigh→max
rename covers legacy tokens, max passes through verbatim.
From the review:
- Rollback safety: persisted reasoning_effort (session summaries, chat
history) deserializes leniently — unknown future tokens degrade to
None with a warning instead of hiding sessions or failing resume.
- Restore migration: a pre-split xhigh override onto a model whose menu
offers max but not xhigh (K3) migrates once, healing display/active-row
drift and re-persisting the live vocabulary.
- /effort max now rejects (with the offered list) on models whose menu
lacks a max row instead of silently applying xhigh; deliberate, tested.
- The interim Responses-backend Max→xhigh downgrade (async-openai has no
Max variant through 0.41) warns loudly; real max wiring lands with the
OpenAI provider cycle via post-serialize body patch.
- Two rusted ignored-e2e wire pins asserted the pre-adapt reasoning_effort
key (deleted by the body adapter since ea0ce9d); they now pin the real
thinking.effort=max shape.
The PRD's first acceptance gate now holds: grep -RinE '\bx\.ai\b|grok'
crates/ --include='*.rs' → 0 matches (exempt: NOTICE and third-party
license archives, README provenance, and the required 'Based on Grok
Build Open Source' attribution, now sourced from version_attribution.txt).
Wire-visible renames (both sides in this repo, changed in lockstep):
- Auth method id 'grok.com' → 'kimi-code' (AuthMethodKind::KimiCode).
- Every x.ai/* and _x.ai/* ACP ext method and meta key → kigi/* /
_kigi/* (~200 names; grokShell → kigiShell). Session-file replay keeps
a read-side alias for the legacy '_x.ai/session/update' method so
existing updates.jsonl histories load; writes emit only the new name
(both directions test-pinned).
- Agent types grok-build* → kigi* with a documented legacy-prefix alias
at resolution time so persisted sessions keep resolving.
- ToolNamespace/BuiltinAgentName GrokBuild* → Kigi* (wire snake_case
kigi/kigi_concise/kigi_hashline; schema regenerated); grok_build
implementation dirs renamed to kigi*.
- x-grok-* headers → x-kigi-*, __GROK_* sentinels → __KIGI_*, themes
grokday/groknight → kigiday/kiginight (old persisted values fall back
to the default theme), web_fetch allowlist xAI hosts → kimi.com +
moonshot platforms, changelog CDN → this repo, grok-build changelog
archives deleted.
- BYOK default endpoint removed: [endpoints] api_base_url is now truly
optional with NO default — consumers fail fast with the flag name when
unset (no silent x.ai egress). Mock harnesses inject it explicitly.
- System-prompt identity fixed: 'released by xAI' → 'an unofficial
community CLI for Kimi' (template + regenerated encrypted form).
Also repaired pre-existing grok-era test debt found by the sweep: the
stale trace_classify default-model pin, the grok-pager UA label test,
pty-harness stale-binary reuse and non-hermetic moonshot routing (a PTY
test could previously reach the real api.moonshot.cn), and the outdated
oauth fixture scope key.
Gates: §9 grep 0; fmt clean; workspace check/clippy 0/0 (-D warnings);
FULL cargo test --workspace: 234 suites, 21,961 passed, 0 failed;
deny advisories ok.
Sampler / inference (PRD F3):
- kimi_compat.rs: single adaptation point for the Kimi chat/completions
dialect (thinking-field mapping, model_id stripping, empty-content
tool-call message fix, stream_options.include_usage), with kimi-cli
source citations
- Rate-limit handling reworked for Kimi/Moonshot semantics; UA kigi/{version}
- /models replaces the xAI models-v2 endpoint everywhere; idle model
refresh carries X-Msh-* device headers only (X-XAI-Token-Auth and
x-grok-client-mode/CLIENT_MODE_HEADER machinery deleted)
Cloud-surface excision (PRD §5, zero-egress):
- remote/ conversations lane, cli-chat-proxy-types crate, prod/ dir,
share command, credit bar: deleted (single local session lane;
paginate() replaces merge_and_paginate)
- Subscription/tier gate stack deleted end-to-end: AppView
gate/tier/team/ZDR fields, app/subscription.rs watch loop,
dispatch/billing.rs paywall + SuperGrok upsell, free-usage-exhausted
chain, tier-restricted commands, GateInfo, RemoteSettings gate fields,
SettingsUpdateNotification gate fields
- /privacy + coding-data-sharing setting deleted (backed by a dead xAI
RPC; Kigi is zero-egress — nothing to share or retain remotely)
Auth UX correctness (user-reported):
- Device-flow fixtures now mirror the live Kimi payload shape
(https://www.kimi.com/code/authorize_device?user_code=..., verified
against auth.kimi.com); the fabricated auth.kimi.com/device?code=...
URLs are gone
- open_browser_detached is a no-op under cfg(test): unit tests drove
wiremock fixture URLs into the real browser (root cause of the
"garbage mock link" ABCD-1234 tabs)
- Welcome/pager-minimal rebrand: Grok Build -> Kigi, grok.com ->
kimi.com, "Sign in to Grok" -> "Sign in to Kimi"
ZacharyZhang-NY