smashimo/oh-my-pi
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix(coding-agent): preserved tool-search history on disk

Browse Source
This commit is contained in:
can1357
2026-08-16 02:31:57 +02:00
parent 544c414296
commit 7af249b190
3 changed files with 30 additions and 10 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/provider-quirks.md
+1 -1
View File
@@ -152,7 +152,7 @@ The Anthropic provider (`packages/ai/src/providers/anthropic.ts`) implements the
- **Claude Code Fingerprint Headers & Betas**: Default headers include `anthropic-version: 2023-06-01`, `anthropic-dangerous-direct-browser-access: true`, `x-app: cli`, and `User-Agent: claude-cli/2.1.220 (external, claude-desktop)` (`coworkUserAgent`). Active beta flags (`buildCoworkBetas`) include `claude-code-20250219`, `interleaved-thinking-2025-05-14`, `thinking-token-count-2026-05-13`, `context-management-2025-06-27`, `prompt-caching-scope-2026-01-05`, `mid-conversation-system-2026-04-07`, `advanced-tool-use-2025-11-20`, `effort-2025-11-24`, and `fallback-credit-2026-06-01` (`context-1m-2025-08-07` is omitted to avoid 429 credit errors on subscription tokens, #7238). Fingerprint metadata (`generateClaudeCloakingUserId`, `deriveClaudeDeviceId`, `generateClaudeJsonUserId`) generates device/session IDs. Billing attestation headers (`createClaudeBillingHeader`, `wrapFetchForCch`, `patchCch`) embed `cch=00000` XXHash64 hashes into `system[0]`.
- **System-Prompt Injection**: `buildAnthropicSystemBlocks` (`packages/ai/src/providers/anthropic.ts`) automatically prepends `claudeCodeSystemInstruction` ("You are a Claude agent, built on Anthropic's Claude Agent SDK.") as `system[0]` for OAuth credentials. Mid-conversation system messages in turn history are enabled for Opus 4.8+ / Sonnet 5+ via `mid-conversation-system-2026-04-07`.
- **Thinking Signatures & Redacted Thinking**: Replaying modified or unsigned thinking blocks causes Anthropic API errors (`invalid signature in thinking block`). `convertAnthropicMessages` converts `ThinkingContent` and `RedactedThinkingContent` (`type: "redacted_thinking"`, `data`). `maybeAddReplayUnsignedThinkingHint` attaches recovery hints on signature errors, while `unwrapAnthropicThinkingEnvelope` strips legacy `<thinking>` XML wrappers.
- **Tool Use Replay & Prefixes**: `encodeAnthropicToolName` / `decodeAnthropicToolName` (`packages/ai/src/providers/anthropic.ts`) prefixes custom tool names with `_` (`claudeToolPrefix`) when using OAuth to prevent collisions with built-in tools (`web_search`, `code_execution`, `text_editor`, `computer`). Server-executed web searches (`ServerToolUseBlockParam`, `WebSearchToolResultBlockParam` in `anthropic-wire.ts`) are detected via `isAnthropicWebSearchHistoryBlock` for turn replay. Empty tool errors are filled by `ensureErrorToolResultWireContent`.
- **Tool Use Replay & Prefixes**: `encodeAnthropicToolName` / `decodeAnthropicToolName` (`packages/ai/src/providers/anthropic.ts`) prefixes custom tool names with `_` (`claudeToolPrefix`) when using OAuth to prevent collisions with built-in tools (`web_search`, `code_execution`, `text_editor`, `computer`). Server-executed web searches and tool searches (`AnthropicServerToolHistoryBlockParam` in `anthropic-wire.ts`) are detected via `isAnthropicServerToolHistoryBlock` for turn replay. Empty tool errors are filled by `ensureErrorToolResultWireContent`.
- **Strict-Tool Schema Normalization & Fallback**: `normalizeAnthropicToolSchema` and `normalizeAnthropicStrictSchema` strip unsupported JSON schema keywords (e.g. `minItems`/`maxItems` on objects) for the `structured-outputs-2025-12-15` beta. If a strict tool schema causes HTTP 400, `streamAnthropicOnce` calls `dropAnthropicStrictTools` and automatically retries without strict mode.
- **Adaptive vs Budget Thinking**: `ThinkingConfigParam` (`anthropic-wire.ts`) supports budget thinking (`{ type: "enabled", budget_tokens: N }` enforced by `ensureMaxTokensForThinking`) and adaptive thinking (`{ type: "adaptive" }` paired with `output_config: { effort: level }` via `effort-2025-11-24` beta). Forced tool choices (`disableThinkingIfToolChoiceForced`) automatically disable thinking.
- **Prompt Cache Breakpoints**: `applyPromptCaching` (`packages/ai/src/providers/anthropic.ts`) attaches `{ type: "ephemeral", scope: "global" }` breakpoints to system prompts (`cacheSystemPrefixBreakpoints`), tool definitions, and historical user turns. `enforceCacheControlLimit` caps total breakpoints to 4 per request.