can1357 0068918f42 fix(ai): dropped scope:global from CC system cache control ...

- Removed global scope field rejected by third-party Anthropic proxies.
- prompt-caching-scope only works against canonical api.anthropic.com.

2026-06-06 14:41:03 +02:00

..

cursor

feat(ai): added conversation caching, JSONL debug logging, and cursor-log.py script

2026-01-11 08:28:09 +01:00

openai-codex

feat: added hideThinkingSummary across stream, agent, session payloads

2026-05-07 23:23:08 +02:00

amazon-bedrock.ts

perf(agent,ai,coding-agent): in-place state mutation + per-delta json-parse throttle + drop hot-path structuredClone

2026-05-31 04:46:06 +02:00

anthropic-client.ts

feat(ai): added internal Anthropic transport with retries and request fallback

2026-06-02 11:04:46 +02:00

anthropic-messages-server-schema.ts

feat(ai): added internal Anthropic transport with retries and request fallback

2026-06-02 11:04:46 +02:00

anthropic-messages-server.ts

feat(ai): added Anthropic task budget support via output_config

2026-06-01 14:35:43 +02:00

anthropic-wire.ts

feat(ai): added internal Anthropic transport with retries and request fallback

2026-06-02 11:04:46 +02:00

anthropic.ts

fix(ai): dropped scope:global from CC system cache control

2026-06-06 14:41:03 +02:00

aws-credentials.ts

fix(ai): support credential_process profiles in AWS credential resolver

2026-05-17 06:56:07 +00:00

aws-eventstream.ts

feat(ai): added AWS auth and Bedrock stream decoding with SigV4

2026-05-17 02:36:43 +02:00

aws-sigv4.ts

feat(ai): added AWS auth and Bedrock stream decoding with SigV4

2026-05-17 02:36:43 +02:00

azure-openai-responses.ts

fix(ai): allow per-model opt-out of max_output_tokens on the wire

2026-06-04 19:18:00 +00:00

cursor.ts

refactor(coding-agent): renamed todo_write tool to todo

2026-06-04 02:45:30 +02:00

error-message.ts

fix: normalized line-ending handling for text and TUI output flows

2026-05-15 23:46:23 +02:00

github-copilot-headers.ts

fix: github.com enterprise routing and structured Copilot OAuth credentials

2026-04-09 18:47:08 +02:00

gitlab-duo.ts

fix(ai/providers): removed hardcoded 32k max token cap

2026-06-04 02:27:23 +02:00

google-auth.ts

feat(ai): added AWS auth and Bedrock stream decoding with SigV4

2026-05-17 02:36:43 +02:00

google-gemini-cli.ts

fix(coding-agent,ai): file-lock race + render-utils sanitization + google named-tool routing

2026-05-31 04:45:57 +02:00

google-gemini-headers.ts

refactor: replaced abortableSleep with scheduler.wait and fetchWithRetry

2026-05-13 16:40:58 +02:00

google-shared.ts

fix(coding-agent,ai): file-lock race + render-utils sanitization + google named-tool routing

2026-05-31 04:45:57 +02:00

google-types.ts

feat(ai): added AWS auth and Bedrock stream decoding with SigV4

2026-05-17 02:36:43 +02:00

google-vertex.ts

fix(providers): discovered vertex project models

2026-05-26 16:34:53 +00:00

google.ts

feat(ai): added AWS auth and Bedrock stream decoding with SigV4

2026-05-17 02:36:43 +02:00

grammar.ts

feat: added compact atom-mode parser and execution support

2026-04-29 01:16:33 +02:00

kimi.ts

refactor: reorganized exports and removed obsolete helper branches

2026-05-14 04:36:19 +02:00

mock.ts

feat(ai/stream-timeouts): added deadline-based first-event timeout handling in stream providers

2026-05-26 15:27:06 +02:00

ollama.ts

fix(agent): handled skipped tool calls for non-toolUse assistant turns

2026-05-29 13:31:37 +02:00

openai-anthropic-shim.ts

fix(ai/providers): removed hardcoded 32k max token cap

2026-06-04 02:27:23 +02:00

openai-chat-server-schema.ts

fix(ai): resolved auth-gateway classification by status-code precedence

2026-05-28 01:45:32 +02:00

openai-chat-server.ts

fix(ai): resolved auth-gateway classification by status-code precedence

2026-05-28 01:45:32 +02:00

openai-codex-responses.ts

fix(ai): kept openai first-event env honored with per-call idle

2026-05-31 19:02:53 +00:00

openai-completions-compat.ts

refactor(ai): moved Kimi K2.6 thinking.keep into compat detection

2026-06-04 15:13:00 +02:00

openai-completions.ts

fix(ai): allow per-model opt-out of max_output_tokens on the wire

2026-06-04 19:18:00 +00:00

openai-responses-server-schema.ts

feat: added auth-gateway usage cache with single-flight 15s ttl fallback

2026-05-17 01:10:25 +02:00

openai-responses-server.ts

chore: reformat

2026-05-30 18:08:51 +02:00

openai-responses-shared.ts

fix(ai): preserved llama parallel tool args

2026-06-06 04:30:05 +00:00

openai-responses.ts

fix(ai): allow per-model opt-out of max_output_tokens on the wire

2026-06-04 19:18:00 +00:00

pi-native-client.ts

feat(ai): added pi-native auth-gateway transport

2026-05-17 03:05:57 +02:00

pi-native-server.ts

feat(ai): added OpenRouter routing-variant suffix support

2026-05-27 18:41:47 +02:00

register-builtins.ts

fix(ai): honored openai idle timeout for ollama streams

2026-06-05 18:36:29 +00:00

synthetic.ts

refactor: reorganized exports and removed obsolete helper branches

2026-05-14 04:36:19 +02:00

transform-messages.ts

feat(mnemosyne): added session-scoped visibility, graph tools, and safety hardening

2026-05-30 15:58:26 +02:00

vision-guard.ts

fix(ai): preserved dashscope compatible-mode multimodal qwen

2026-06-04 13:19:16 +00:00

xai-responses.ts

fix(ai): gate reasoning.effort on Grok allowlist via omitReasoningEffort

2026-05-27 15:42:45 +00:00