Skip to content

CHANGELOG.md

Latest commit

 

History

History
94 lines (77 loc) · 2.59 KB

CHANGELOG.md

File metadata and controls

94 lines (77 loc) · 2.59 KB
Raw

Changelog

All notable changes to this project will be documented in this file.

The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.

[0.1.0] - 2025-12-18

Added

Core Features

  • CLI interface with commands: run, dry-run, status, reset, smoke
  • YAML-based configuration with Pydantic validation
  • Structured JSONL logging with summary statistics
  • SQLite-based idempotency (external_state mode)
  • Tag-based idempotency support (requires custom field)

Salesforce Integration

  • Opportunity querying with configurable filters
  • Event (meeting) creation
  • Task (email) creation
  • Contact lookup
  • Record tagging for cleanup

Activity Generation

  • Deterministic activity planning with seeded RNG
  • Configurable past/future time windows
  • Realistic meeting and email subjects
  • Participant role assignment

Content Generation

  • OpenAI integration for realistic content
  • Meeting notes generation
  • Email body generation
  • Scorecard answer generation
  • Heuristic fallback when LLM disabled

Scorecard Support

  • MEDDICC template implementation
  • Hybrid mode (LLM + heuristic)
  • Confidence scoring
  • Coverage tracking

Documentation

  • Comprehensive README
  • Quick start guide
  • Architecture documentation
  • Example configuration
  • Unit tests

Technical Details

  • Python 3.9+ support
  • Click CLI framework
  • Rich terminal output
  • Makefile for common tasks
  • VSCode configuration

[Unreleased]

Added

  • Thread-safe logging and state store for concurrent runs
  • Opportunity-level concurrency via --concurrency
  • Tag-mode run tagging on Event/Task creation
  • Reset cleanup for tag mode and external_state using state.sqlite
  • Config validation for date ranges and min/max bounds
  • Automatic .env loading in CLI

Changed

  • Scorecard scoring now uses template question counts
  • Scorecard generation guarantees at least one answer when coverage_target > 0
  • SOQL generation now escapes string literals and quotes dates

Planned for 0.2.0

  • Progress bars during runs
  • People.ai API verification
  • Additional scorecard templates (BANT, MEDDPICC)
  • Custom activity subjects via configuration
  • Better error recovery and retry logic

Planned for 0.3.0

  • Web UI for configuration
  • Scheduled runs
  • Diff mode for state comparison
  • Export/import state functionality
  • Cloud deployment options

Future Considerations

  • Contact auto-generation
  • Account hierarchy support
  • Multi-org orchestration
  • Reporting dashboard
  • Slack/Teams notifications
  • Webhook support for integrations