Introduce `anyllm_client` crate containing HTTP client construction,
SSRF-safe DNS resolver, retry/backoff logic, rate limit header parsing,
and SSE frame parsing. These were previously inlined in the proxy crate.
Rename crates from `anthropic_openai_proxy`/`anthropic_openai_translate`
to `anyllm_proxy`/`anyllm_translate` throughout.
proxy/backend: now re-exports retry, rate limit, and SSE symbols from
the client crate; `send_with_retry` and `build_http_client` are thin
adapters bridging BackendAuth/TlsConfig to the client crate's types.
streaming: remove duplicate `find_double_newline` and
`MAX_SSE_BUFFER_SIZE` definitions; import from `crate::backend` instead.
Fix missing `pub mod` declarations in translator and proxy that were
accidentally replaced by doc comments (streaming, usage_map, server,
redact).
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Add doc comments to ~30 undocumented public functions, structs, enum
variants, and fields across both crates. Strengthen weak comments to
explain "why" not just "what". Remove 16 stale PLAN.md line references
(PLAN.md was removed previously). Add Anthropic API doc links where
relevant for rate limit headers and ID format.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Handle o-series reasoning models (o1, o3, o4-mini): drop max_tokens
(keep only max_completion_tokens) and convert system -> developer role
- Restore read_timeout(900s) alongside tcp_keepalive(60s) to bound hung
connections that keepalive alone cannot detect
- Reuse epoch_to_iso8601 from admin/db.rs instead of duplicating the
Hinnant civil date algorithm in backend/mod.rs
- Make ISO 8601 conversion testable via anchor time parameter
- Fix f64 truncation in duration parsing (use .round() before cast)
- Extract convert_reset_duration helper to deduplicate header injection
- Use eq_ignore_ascii_case in is_o_series_model to avoid allocation
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Security: hash API keys with SHA-256 for constant-time comparison (eliminates
length timing leak), require explicit PROXY_OPEN_RELAY for unauthenticated
access, gate /metrics behind auth, sanitize admin error responses, enforce
log_level allowlist on DB restore, validate GCP identifiers against URL
injection, add referrer-policy header to admin SPA.
Correctness: hold concurrency semaphore permit through entire stream lifetime
(not just until headers are sent), fix SSE parser to resume scanning near
chunk boundaries instead of re-scanning from start, mark padding tool call
slots as closed to prevent spurious ContentBlockStop events, mark responses
streaming translator as finished on error to prevent double closure events,
serialize concurrent config writes with a Mutex.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Previously, backend errors (401, 429, 500) that occurred before any SSE
data was sent were wrapped in a 200 OK response with an error event
buried in the stream. Now messages_stream returns Result so the caller
can respond with the correct HTTP status code.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Prevents hard deserialization failures from unknown finish reasons
(e.g. DeepSeek's "insufficient_system_resource") and adds bidirectional
mapping of reasoning_content to/from Anthropic thinking blocks for
DeepSeek/Qwen thinking models.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Split config.rs into config/{mod,tls,url_validation}.rs and extract
server/{passthrough,streaming,token_counting}.rs and admin/ws.rs from
their parent modules for clarity.
Functional changes:
- Add DNS rebinding protection on admin API (Host header validation)
- Fix IPv6 host parsing in admin origin check (bare ::1 was mishandled)
- Return Anthropic-shaped errors for JSON parse failures (400 not 422)
and unmatched routes (404 not_found_error)
- Forward extra fields from Anthropic request to OpenAI request
- Count "Error: " prefix tokens for error tool results
- Reduce TOKENIZER visibility to module-private
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Use constant-time comparison (subtle::ConstantTimeEq) for proxy API key
validation, matching the pattern already used in admin auth
- Restrict admin config log_level to allowlist (error/warn/info/debug) to
prevent trace-level logging that leaks secrets in HTTP headers
- Create admin token file with atomic 0o600 permissions via OpenOptionsExt
to eliminate TOCTOU race where file is briefly world-readable
- Add .admin_token, *.db, and TLS key material to .gitignore
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Constant-time token comparison in admin auth (timing side-channel)
- Origin header parsing via URL to prevent bypass (e.g., 127.0.0.1.attacker.com)
- CSP headers and X-Frame-Options on admin SPA
- WebSocket origin check for cross-site WS hijacking prevention
- Switch admin DB mutex from tokio::Mutex to std::Mutex, use spawn_blocking
- Poison-recovery on std::sync locks (unwrap_or_else + into_inner)
- JSON builder for SSE error fallback to prevent injection
- UTF-8-safe secret redaction (char-aware slicing)
- Add RedactedThinking content block and SignatureDelta streaming types
- Map OpenAI refusal field to Anthropic text block (non-streaming + streaming)
- Map OpenAI cached_tokens to Anthropic cache_read_input_tokens
- Map HTTP 408 to Anthropic OverloadedError
- Warn on trace log level and log_bodies enable via admin API
- Update CLAUDE.md docs to match current state
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Gemini's OpenAI-compatible endpoint (/openai) supports the same Chat
Completions format as the OpenAI backend, making the native Gemini
translation path redundant. This removes ~3400 lines of Gemini-specific
types, mapping, streaming, and client code, routing Gemini through the
existing OpenAI translation pipeline instead.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Security: strip admin token from browser URL, add cross-origin rejection
middleware for admin API, skip tool calls with empty names instead of
substituting "unknown".
Correctness: use BytesMut for SSE buffering to prevent UTF-8 corruption
at TCP chunk boundaries, use saturating_sub for epoch arithmetic, handle
CRLF SSE frame delimiters.
Runtime: switch runtime_config to std::sync::RwLock (guard is !Send),
use block_in_place for SQLite IO, spawn_blocking for tokenization,
add tracing reload layer so admin log_level changes apply immediately.
Reliability: retry failed log buffer flushes with capped retry queue,
add MAX_SSE_BUFFER_SIZE guard in middleware handler, cap tool call and
part indices to prevent unbounded vec growth.
Observability: defer streaming request logging until stream completes
so entries capture actual status, latency, and token counts.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
README: rewrite with use cases, Claude Code / local LLM / OpenRouter
examples, multi-backend TOML config, admin dashboard documentation,
and updated endpoint table.
CLAUDE.md: update test counts, add Anthropic backend, PROXY_API_KEYS,
admin module, and correct architecture references.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- PROXY_API_KEYS env var: comma-separated allowlist of valid API keys.
When set, only listed keys are accepted. When unset, any non-empty key
is accepted (backward-compatible behavior).
- DNS resolution validation: base URL validation now resolves domain
names at startup and rejects any that resolve to private/loopback IPs,
hardening against DNS rebinding attacks.
- Made is_private_ip public for reuse.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The previous commit added `pub mod anthropic_client` to backend/mod.rs
and wired up BackendKind::Anthropic in config and routes, but did not
include the actual client implementation file -- breaking compilation.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Wire up Responses API as alternative to Chat Completions via
OPENAI_API_FORMAT=responses env var. Adds request/response mapping,
streaming state machine, client methods, and route dispatch. Includes
live API integration tests (ignored by default, require OPENAI_API_KEY).
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Collapse retry + final-attempt into single inclusive loop, drain response
body before retry to return connections to pool, and replace panic on
invalid x-request-id with UUID fallback. Update CLAUDE.md/TASKS.md for
Phase 22 status. Apply cargo fmt across touched files.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Pure type definitions for Gemini generateContent API: request/response
structs, Content/Part union, tool declarations, error shapes, safety
settings, and citation metadata. 19 serde round-trip tests with golden
fixtures. No mapping logic or proxy changes.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Phase 19: Replace count_tokens stub with tiktoken-rs implementation using
o200k_base encoder. Extracts text from system prompt, messages, tool
definitions, and thinking blocks for approximate token counting.
Phase 20b: Extract BackendClient enum in backend/mod.rs with OpenAI/Vertex
variants and dispatch methods, preparing for future native Gemini backend.
Pure refactor with no behavior change.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Parameterize OpenAIClient with BackendAuth enum to support both OpenAI
and Vertex AI backends. BACKEND=vertex enables Vertex AI with
VERTEX_PROJECT, VERTEX_REGION, and VERTEX_API_KEY/GOOGLE_ACCESS_TOKEN
env vars. Model defaults change to gemini-2.5-pro/gemini-2.5-flash
for Vertex. No new client type; reuses OpenAIClient with different
URL construction and auth header dispatch.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Phases 12-18: release infrastructure (LICENSE, README, Dockerfile, CI,
CHANGELOG), transparent proxy with anthropic-version/anthropic-beta header
passthrough and lossy translation warnings, BIG_MODEL/SMALL_MODEL env-based
model mapping, mTLS client cert support (P12/PEM), max_completion_tokens
and reasoning_effort passthrough via serde flatten, extended thinking type
support (thinking blocks stripped in translation), top_k typed field.
Phase 20: Gemini backend research with docs/gemini-api-diffs.md covering
native API format, tool calling, streaming, auth, schema restrictions,
and Vertex AI OpenAI-compatible endpoint. Task roadmap through Phase 22.
Test count: 169 -> expanded with new fixture and unit tests for all phases.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Rust workspace with two crates:
- translator: pure, IO-free mapping between Anthropic Messages API and OpenAI Chat Completions
- proxy: axum HTTP server with auth, streaming SSE, retry/backoff, concurrency limits
169 tests passing (unit, golden fixture, integration).
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
