Skip to content

Commit 87a876a

Browse files
committed
fix(sync): align stale version refs to v0.9.6.0, fix CHANGELOG date typo
- .env.example: v0.9.4.3 → v0.9.6.0 header - hardware/SKILL.md: v0.9.4.3 → v0.9.6.0 - CHANGELOG.md: fix 2025 → 2026 date typo in [0.9.5.0] - CHANGELOG.md: add missing [0.9.6.0] entry - No version bumps — aligning existing references only Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
1 parent 7a7dfb6 commit 87a876a

3 files changed

Lines changed: 24 additions & 3 deletions

File tree

.env.example

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1,4 +1,4 @@
1-
# Perplexity-Tools v0.9.4.3 Configuration
1+
# Perplexity-Tools v0.9.6.0 Configuration
22
# Copy to .env and fill in your actual values
33

44
# === PERPLEXITY API (Cloud Models) ===

CHANGELOG.md

Lines changed: 22 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -8,7 +8,28 @@ Cross-repo changes affecting ultrathink-system are marked with `[SYNC]`.
88

99
---
1010

11-
## [0.9.5.0] - 2025-03-26
11+
## [0.9.6.0] - 2026-03-27
12+
13+
### Added
14+
- **LAN Continuity**: LAN Detect & Resume for seamless multi-computer operation [SYNC]
15+
- **Spawn Reconciliation**: Pre-flight spawn detection and reconciliation before model spawning [SYNC]
16+
- **Short Persistence Log**: `.state/session.log` for low-overhead session tracking
17+
- `orchestrator/lan_discovery.py` — LAN-wide AI model discovery
18+
- `orchestrator/spawn_reconciliation.py` — ECC and autoresearch spawn reconciliation registry
19+
- `tests/test_lan_discovery.py` — LAN discovery test coverage
20+
21+
### Changed
22+
- SKILL.md updated to v0.9.6.0 with hardware-aware multi-computer orchestration
23+
- Models updated to Qwen 3.5 series (9B MLX on Mac, 35B MoE on Dell)
24+
- `orchestrator.py` hardened with VRAM safety rules and hardware-bound routing
25+
- Adapted durable workflow and intelligent routing for multi-computer LAN [SYNC]
26+
27+
### Synced with ultrathink-system
28+
- Both repos synchronized to v0.9.6.0
29+
- ultrathink `api_server.py` updated to v0.9.6.0 with GPU reconciliation
30+
- Cross-repo SKILL.md references established for recursive sub-skill loading
31+
32+
## [0.9.5.0] - 2026-03-27
1233

1334
### Added
1435

hardware/SKILL.md

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -2,7 +2,7 @@
22
# hardware/SKILL.md
33
# Decoupled hardware profiles for role-based agent assignment.
44
# Synchronized across all repos. Governs VRAM/RAM limits for model routing.
5-
# Last updated: 2026-03-27 | Version: 0.9.4.3
5+
# Last updated: 2026-03-28 | Version: 0.9.6.0
66

77
---
88

0 commit comments

Comments
 (0)