v1.7.0

v1.7.0 (2026-01-09)

This release is published under the MIT License.

Bug Fixes

• image_prompt_agent: Handle missing fallback_tracker before using it (99db9c6)

Documentation

• copilot-instructions: Document multi-run usage, PowerShell guidance, and CLI options (d0422d0)

• multi-run: Add MULTI_RUN_FEATURE.md summarizing implementation, tests, and usage (a2d18ad)

• readme: Add multi-run usage, behavior, and example output (56e21ad)

Features

• cli: Add --runs option and multi-run sequential execution (32a9bc6)

• database: Add helpers to fetch used topics and bulk insert new topics (5d28862)

• orchestrator: Accept run_number for multi-run runs and record posted topics (c18c7e3)

• run_context: Support optional run_number for multi-run folder naming (f4cfe7c)

• topic_agent: Add LLM batch topic generation and DB replenishment (4a02459)

---

Detailed Changes: v1.6.0...v1.7.0

v1.6.0

v1.6.0 (2025-12-31)

This release is published under the MIT License.

Features

• prompt-generator: Enforce template headings and add self-repair pass; aggregate token usage (9fea4b6)

---

Detailed Changes: v1.5.0...v1.6.0

v1.5.0

v1.5.0 (2025-12-20)

This release is published under the MIT License.

Documentation

• github: Update first-time setup wording to accept any custom field (fd8c220)

• readme: Clarify field selection and examples for freeform fields (6c9ca8c)

Features

• config: Allow arbitrary custom field input in CLI and config (4ae2c19)

---

Detailed Changes: v1.4.0...v1.5.0

v1.4.0

v1.4.0 (2025-12-14)

This release is published under the MIT License.

Bug Fixes

• agents/image-generator: Simplify print statements in standalone test output (e276882)

• core/llm-clients: Use correct response_modalities value for image generation (0e495b0)

• core/llm_clients: Format note for clarity in image generation response modalities (d5d1f4c)

• cost_tracking: Accept keyword args in record_call for backwards-compatible API (b628570)

• cost_tracking: Make agent_name_or_model optional to support pure keyword calls (9f9a1c1)

• orchestrator: Record agent failure metrics when circuit breaker trips\n\nAdd explicit except CircuitBreakerTrippedError in _execute_agent_with_retry() to log duration and error details before propagating. This ensures un_failed.json and metrics capture the failed step even if the breaker opens mid-execution. (a131de0)

• retry: Treat Gemini quota/rate-limit errors as non-retryable to avoid futile retries and circuit breaker trips\n\nAdd _is_quota_or_rate_limit_error() to detect RESOURCE_EXHAUSTED/quota exceeded/rate limit messages. When detected, flip etryable to false and surface actionable guidance. This prevents exponential backoff from hammering the API and stops breaker from accumulating failures on quota issues.\n\nImpact:\n- More graceful aborts under quota exhaustion\n- Clearer error messaging for remediation\n- Preserves circuit breaker for genuine transient failures (1055678)

• reviewer: Scrub forbidden newsletter mentions and improve review pipeline (c42e5eb)

• reviewer: Use positional args when recording cost and include agent name (a30c6fa)

• writer: Remove newsletter sign-off and scrub forbidden mentions (c7efe0d)

• writer_agent: Accept structured_prompt string and use it as LLM prompt (f827069)

Features

• agents/image-generator: Capture full tracebacks on ModelError and add standalone CLI for testing (b608511)

---

Detailed Changes: v1.3.0...v1.4.0

v1.3.0

v1.3.0 (2025-12-13)

This release is published under the MIT License.

---

Detailed Changes: v1.2.0...v1.3.0

v1.2.0

v1.2.0 (2025-12-13)

This release is published under the MIT License.

Bug Fixes

• Resolve flake8 line length issue in main.py (1224284)

• orchestrator: Omit image artifact in run summary when --no-image is set (2717d07)

Documentation

• Add --no-image flag documentation and cost optimization section (cdab745)

• readme: Update README.md to clarify command line execution and mark phases 11 and 12 as complete (9f069b5)

Features

• Add --no-image CLI flag to skip image generation and reduce costs (11c6dce)

• Implement no_image parameter in orchestrator for cost optimization (b882840)

• dry-run: Implement dry-run mode for cost estimation and setup verification (afd2588)

---

Detailed Changes: v1.1.0...v1.2.0

v1.1.0

v1.1.0 (2025-12-07)

This release is published under the MIT License.

Documentation

• phase12: Update ROADMAP with user approval system enhancements (f73ad77)

Features

• Improve fallback resilience and test infrastructure (46ee7d1)

• phase12: Implement user approval system for fallback workflows with FallbackTracker (9098147)

---

Detailed Changes: v1.0.3...v1.1.0

v1.0.3

v1.0.3 (2025-11-29)

This release is published under the MIT License.

Bug Fixes

• roadmap: Mark Conventional Commits enforcement as completed (186f04e)

---

Detailed Changes: v1.0.2...v1.0.3

v1.0.2

v1.0.2 (2025-11-29)

This release is published under the MIT License.

Bug Fixes

• errors: Pass all args to super().init and override str for B042 (099a764)

---

Detailed Changes: v1.0.1...v1.0.2

v1.0.1

v1.0.1 (2025-11-29)

This release is published under the MIT License.

Bug Fixes

• Pass only message to Exception.init in BaseAgentError (efd9893)

• lint: Task 2 bugbear fixes (B042, B014, B011, B017) (c497f7c)

---

Detailed Changes: v1.0.0...v1.0.1