Skip to content

CHANGELOG.md

Latest commit

 

History

History
24 lines (18 loc) · 1.43 KB

CHANGELOG.md

File metadata and controls

24 lines (18 loc) · 1.43 KB
Raw

Changelog

0.2.0 (2026-04-17)

Telemetry + cost attribution parity with @sidclaw/sdk@0.1.11.

Added:

  • record_telemetry(trace_id, params) method on SidClaw and AsyncSidClaw — PATCH /api/v1/traces/:id/telemetry. Token usage and cost are accumulated server-side; outcome_summary and model are set-once (first write wins).
  • RecordTelemetryParams TypedDict with tokens_in, tokens_out, tokens_cache_read, model, cost_estimate, outcome_summary (all optional).
  • Extended RecordOutcomeParams with eight optional fields: outcome_summary, error_classification, exit_code, tokens_in, tokens_out, tokens_cache_read, model, cost_estimate.
  • ErrorClassification literal type: 'timeout' | 'permission' | 'not_found' | 'runtime'.
  • New module sidclaw.cost — MODEL_PRICING table for 13 models (Claude 4.x, GPT-4o, Gemini), estimate_cost(model, tokens_in, tokens_out, tokens_cache_read), register_model_pricing(model, pricing) for user overrides.

No breaking changes — all new fields are optional.

0.1.0 (2026-03-23)

  • Initial release
  • Sync and async clients (SidClaw, AsyncSidClaw)
  • Policy evaluation, approval polling, outcome recording
  • Framework integrations: LangChain, CrewAI, OpenAI Agents, Pydantic AI
  • MCP governance proxy with CLI (sidclaw-mcp-proxy)
  • Webhook signature verification
  • Typed errors: ActionDeniedError, ApprovalTimeoutError, ApprovalExpiredError, RateLimitError