Add cache_control ephemeral breakpoints on system prompt and tool
definitions to activate Anthropic prompt caching for multi-turn
agent loops. Fix token double-counting by removing the redundant
totalTokensUsed accumulator and computing it from sub-items. Surface
cache read/write statistics in both text and JSON summary output.
Migrated internal/llm from manual net/http implementation to anthropic-sdk-go v1.47.0
and openai-go/v3 v3.39.0, reducing ~400 lines of retry, request-building, and streaming
code. Simplified LLMClient interface to single CompletionsWithCtx method. Fixed Anthropic
auth to use WithAuthToken (Bearer) instead of WithAPIKey (X-Api-Key) for proxy compatibility.
Fixed OpenAI message builder to use ExtractText() for proper []ContentBlock handling.
StreamCompletion used http.NewRequest without context and retry with
context.Background(), making streaming calls uncancellable. Memory
compression called Completions() without context, leaving async
goroutines unable to be cancelled or timed out.
- Add StreamCompletionWithCtx to LLMClient interface and both
OpenAI/Anthropic implementations
- Propagate context through runCompression, addNextMessage, and
triggerAsyncCompression call chain
- Add 5-minute timeout cap on async compression goroutines
- Cancel in-flight compression HTTP requests when job is superseded
Allow users to configure either a base URL (e.g. https://api.example.com/v1)
or a full endpoint URL for OpenAI-compatible services. The client now
normalizes the URL on initialization, matching the existing behavior of
the Anthropic client.
Unify the config folder name to `.opencodereview` across all runtime paths,
tests, docs, and i18n strings. Also make defaultConfigPath() return an error
instead of silently falling back to a current-directory file when $HOME is
unresolvable.
kite