helloagents 2.2.15 → 2.2.16

package/README.md

@@ -8,7 +8,7 @@
 
 **A multi-CLI workflow system that keeps going until tasks are implemented and verified.**
 
-[![Version](https://img.shields.io/badge/version-2.2.15-orange.svg)](./pyproject.toml)
+[![Version](https://img.shields.io/badge/version-2.2.16-orange.svg)](./pyproject.toml)
 [![npm](https://img.shields.io/npm/v/helloagents.svg)](https://www.npmjs.com/package/helloagents)
 [![Python](https://img.shields.io/badge/python-%3E%3D3.10-3776AB.svg)](./pyproject.toml)
 [![Commands](https://img.shields.io/badge/workflow_commands-15-6366f1.svg)](./helloagents/functions)
@@ -337,7 +337,7 @@ To install from the `beta` branch, append `@beta` to the repository URL:
 
 - AGENTS.md: router and workflow protocol
 - SKILL.md: skill discovery metadata for CLI targets
-- pyproject.toml: package metadata (v2.2.15)
+- pyproject.toml: package metadata (v2.2.16)
 - helloagents/cli.py: installer entry
 - helloagents/functions: workflow commands
 - helloagents/stages: design, develop
@@ -402,14 +402,18 @@ These commands run inside AI chat, not your system shell.
 
 ## Version History
 
-### v2.2.15 (current)
-
-- Professional evaluation dimensions: Scope(0-3), Specs(0-3), Constraints(0-2), Acceptance(0-2), mapped to PM fundamentals
-- Evaluation threshold raised from 7 to 8, ensuring all zero-score dimensions are covered during follow-up
-- Dimension isolation rule: each follow-up question targets exactly one dimension, no cross-dimension merging
-- Simplified option generation rules, removed redundancy with dimension isolation
-- R3 proposal differentiation strengthened: both technical path and deliverable design direction must differ
-- Universal task support: terminology generalized beyond programming to cover all task types (documents, design, general tasks)
+### v2.2.16 (current)
+
+- Refactored evaluation dimension system with dimension isolation rule, pass threshold tuned to 8/10. Options are user-need-driven, organized by style direction rather than complexity tiers (e.g. UI design offers different styles instead of simple/medium/complex); recommended option points to the most complete deliverable, derived from recommendation principles and scoring criteria rather than hardcoded
+- Proposal design requires both implementation path and deliverable design direction to differ across alternatives, each sub-agent independently outputs a complete proposal including presentation direction, style, and experience
+- Proposal evaluation criteria optimized: user value weight is always no less than any other single dimension, evaluation dimensions dynamically adjusted by project context
+- Universal task type support: generalized evaluation, follow-up, and proposal design terminology from programming-specific to documents, design, general tasks and more
+- Added sub-agent DAG dependency scheduling with topological sort, layer-by-layer parallel dispatch, and failure propagation
+- Dynamic sub-agent parallel count based on independent work units, eliminated hardcoded limits
+- Unified output format: structured display for score breakdowns, follow-up options, and confirmation messages
+- Streamlined execution paths with shorter stage chains and step-level on-demand module loading
+- Adjusted Codex CLI memory limit to 128 KiB to prevent rules file truncation
+- Improved recommendation option generation rules, proposal differentiation requirements, and evaluation scoring criteria
 
 ### v2.2.14
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "helloagents",
-  "version": "2.2.15",
+  "version": "2.2.16",
   "type": "module",
   "description": "HelloAGENTS - AI-native sub-agent orchestration framework for multi-CLI environments",
   "author": "HelloWind",