Commit Graph

9 Commits

Author SHA1 Message Date
Zhaoyikaiii 1f7cbd9164 fix: cache system prompt with mtime-based auto-invalidation (#607)
Avoid rebuilding the entire system prompt on every BuildMessages() call
by caching the static portion (identity, bootstrap, skills summary,
memory) and only recomputing it when workspace source files change.

Key changes:

- ContextBuilder caches the static prompt behind an RWMutex with
  double-checked locking. Source file changes are detected via cheap
  os.Stat mtime checks so no explicit invalidation is needed.

- Track file existence at cache time (existedAtCache map) so that
  newly created or deleted bootstrap/memory files also trigger a
  rebuild — the old modifiedSince() silently returned false on
  os.IsNotExist.

- Walk the skills directory recursively with filepath.WalkDir to
  catch content-only edits at any nesting depth; directory mtime
  alone misses in-place file modifications on most filesystems.

- ToolRegistry.sortedToolNames() sorts tool names before iteration,
  ensuring deterministic tool definition order across calls — a
  prerequisite for LLM-side prefix/KV cache reuse.

- Merge all context (static + dynamic + summary) into a single
  system message for provider compatibility: the Anthropic adapter
  extracts messages[0] as the top-level system parameter, and Codex
  reads only the first system message as instructions.

- Fix a data race in BuildMessages() where cachedSystemPrompt was
  read without holding the lock in a debug log statement.

- Add tests: single system message invariant, mtime auto-invalidation,
  new-file creation detection, skill file content change, explicit
  InvalidateCache, cache stability, concurrent access (20 goroutines
  x 50 iterations, passes go test -race), and a benchmark.
2026-02-25 15:27:45 +08:00
Lixeer d09c64fcee fix: implement code review suggestions
Address all feedback from PR review:
- Lock granularity
- Empty response handling
- Shutdown race condition
- Interface naming
2026-02-24 22:33:04 +08:00
Lixeer a849e02917 fix: better session management for github_copilot_provider 2026-02-22 22:30:53 +08:00
Artem Yadelskyi 0675ce7c38 feat(fmt): Fix formatting 2026-02-20 20:03:11 +02:00
yinwm 7f241647be feat(providers): add thought_signature support for gemini
Add support for persisting thought_signature metadata from Google/Gemini 3
models. This introduces ExtraContent and GoogleExtra types to handle
provider-specific metadata, and ensures thought signatures are properly
preserved through the tool call lifecycle.
2026-02-20 00:36:31 +08:00
Leandro Barbosa 447c17aeb1 merge: sync upstream/main (PR #213) into feat/multi-agent-routing
Resolve conflicts in pkg/providers/types.go and pkg/agent/loop.go:
- types.go: use protocoltypes aliases from PR #213, keep fallback types
- loop.go: drop old single-agent createToolRegistry (replaced by multi-agent pattern)

Refactor to align with PR #213 patterns:
- instance.go: use NewExecToolWithConfig (accept full config for deny patterns)
- registry.go: pass full config to NewAgentInstance
- loop.go: add Perplexity web search options to registerSharedTools
2026-02-18 11:39:14 -03:00
Jared Mahotiere c4cbb5fb35 providers: finalize PR213 review fixes
Phase 1: centralize protocol message/tool/response types in protocoltypes and keep compatibility aliases in providers and protocol packages.

Phase 1: preserve HTTPProvider constructor compatibility and route Anthropic api_base through factory auth/provider constructors with base URL normalization.

Phase 2: expand provider routing/auth tests (deepseek/nvidia/shengsuanyun, codex/claude oauth/codex-cli) and add openai_compat + anthropic coverage for proxy transport, model normalization, numeric option coercion, token-source refresh, and base URL behavior.

Phase 3: apply gofmt and validate with Dockerized tests (go test ./pkg/providers/... ./pkg/migrate and go test ./...).
2026-02-17 11:13:10 -05:00
Leandro Barbosa 6e7149509a feat: add model fallback chain with error classification
Add 2-layer fallback system (text + image) with automatic candidate
resolution. Includes error classifier (~40 patterns), per-provider
cooldown (exponential backoff), and model reference parsing.

- FailoverError/FailoverReason types for structured error handling
- ErrorClassifier with rate_limit, billing, auth, timeout patterns
- FallbackChain with cooldown management and candidate rotation
- ModelRef parser for provider/model string format
- 128 tests, 95%+ coverage
2026-02-13 12:12:12 -03:00
lxowalle e17693b17c * First commit 2026-02-09 19:20:19 +08:00