Releases: NickCrew/Claude-Cortex
v4.4.3
v4.4.3 (2026-04-18)
This release is published under the MIT License.
Bug Fixes
-
agent-loops: Add polling, heartbeats, and failure diagnostics to review scripts (
e4b1a1c) -
agent-loops: Hard-fail oversized review inputs instead of silent truncation (
36a41e2)
Detailed Changes: v4.4.2...v4.4.3
v4.4.2
v4.4.2 (2026-04-14)
This release is published under the MIT License.
Bug Fixes
- tmux: Rename send positional to avoid clobbering top-level dispatcher (
db9bd2f)
Continuous Integration
Detailed Changes: v4.4.1...v4.4.2
v4.4.1
v4.4.1 (2026-04-13)
This release is published under the MIT License.
Bug Fixes
Documentation
-
reference: Publish skill recommendation system deep dive to site (
b6c0b67) -
skills: Add working-with-skills to guides landing list (
c6f1415)
Detailed Changes: v4.4.0...v4.4.1
v4.4.0
v4.4.0 (2026-04-12)
This release is published under the MIT License.
Chores
- hooks: Remove legacy plugin manifest and stale moved-readme (
43c1f75)
Documentation
-
skills: Add working-with-skills guide and rewrite hook install story (
29fde1d) -
tutorials: Publish skill authoring tutorial to site (
228e6de)
Features
- skills: Thread prompt text through skill recommender (
8718245)
Detailed Changes: v4.3.1...v4.4.0
v4.3.1
v4.3.1 (2026-04-12)
This release is published under the MIT License.
Bug Fixes
- agent-loops: Detect gemini review sessions (
b4a8e41)
Documentation
- agent-loops: Mark review/audit scripts as blocking calls with timeout guidance (
4a5d8cf)
Detailed Changes: v4.3.0...v4.3.1
v4.3.0
v4.3.0 (2026-04-07)
This release is published under the MIT License.
Bug Fixes
-
agent-loops: Add prominent source-only scoping rule near invocation reference (
c4b3590) -
agent-loops: Clarify specialist-review scopes to source files only (
8156173) -
agent-loops: Clear parent session env vars for nested codex exec (
decd517) -
agent-loops: Enforce mandatory backlog filing for deferred findings (
214493e) -
agent-loops: Expand test directory auto-discovery and fix truncation bug (
af7109a) -
agent-loops: Improve review script reliability from sandboxed contexts (
74364b5) -
agent-loops: Reject test files in specialist-review path filter (
7e1fc69) -
agent-loops: Remove --tools "" flag that broke Claude CLI invocation (
5bd0116)
Features
-
cli: Add unified cortex suggest command (
20732d1) -
skills: Add 7 community skills from claude-skills library (
c71a593) -
skills: Add business, legal, and research skills from claude-skills (
4453f37) -
skills: Add copywriter, email-drafter, and proofreader skills (
72a61e0) -
skills: Add data visualization, design, and UX writing skills (
9b083ca) -
skills: Add storyteller skill for brand narratives and story structure (
c12d9ca)
Detailed Changes: v4.2.0...v4.3.0
v4.2.0
v4.2.0 (2026-04-05)
This release is published under the MIT License.
Features
Detailed Changes: v4.1.1...v4.2.0
v4.1.1
v4.1.1 (2026-04-05)
This release is published under the MIT License.
Bug Fixes
-
agent-loops: Add guard preventing non-Claude agents from attempting team review (
598e1ca) -
agent-loops: Direct non-Claude agents to escalate team review to user (
d87c06a)
Detailed Changes: v4.1.0...v4.1.1
v4.1.0
v4.1.0 (2026-04-05)
This release is published under the MIT License.
Documentation
- agent-loops: Add teams-based multi-specialist review mode to SKILL.md (
bc234f7)
Features
- agent-loops: Add multi-specialist team review scripts and templates (
d3e57d2)
Detailed Changes: v4.0.1...v4.1.0
v4.0.1
v4.0.1 (2026-04-05)
This release is published under the MIT License.
Bug Fixes
-
build: Resolve TOML syntax errors and duplicate project.urls (
4a1bd92) -
ci: Switch PyPI publishing to Trusted Publisher (OIDC) (
7bf3a7d) -
test: Handle auth status probe in specialist review test (
14a787e)
Detailed Changes: v4.0.0...v4.0.1