@@ -7,6 +7,55 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
77
88## [ Unreleased]
99
10+ ## [ 0.7.0] - 2025-01-27
11+
12+ ### Added
13+ - ** Continue CLI Agent Support**
14+ - New adapter for Continue CLI (@continuedev/cli ) - AI coding assistant with TUI and headless modes
15+ - Command: ` cn ` (not ` continue ` )
16+ - Installation via npm: ` npm install -g @continuedev/cli `
17+ - Version tested: 1.5.12
18+ - Key features:
19+ - ` -p ` flag for prompt input (different pattern from Claude/Gemini's stdin approach)
20+ - ` --model ` flag for model specification (e.g., gpt-4, claude-sonnet-4-5)
21+ - ` --silent ` flag to strip ` <think></think> ` tags from output
22+ - Authentication via ` cn login ` to Continue Mission Control
23+ - Full Agent interface implementation with structured prompts
24+ - Message filtering to prevent echo in multi-agent conversations
25+ - Streaming and non-streaming message support
26+ - Example configurations:
27+ - ` examples/continue-coding.yaml ` - Single Continue agent for refactoring
28+ - ` examples/continue-team-coding.yaml ` - Multi-agent team (Continue + Claude + Gemini)
29+ - Complete test coverage and documentation
30+ - Now supporting 16 AI agent CLIs (up from 15)
31+
32+ ### Technical Details
33+ - ** New Module** : ` pkg/adapters/continue.go `
34+ - Flag-based prompt pattern: ` cn -p "prompt" ` instead of stdin
35+ - Implements structured three-part prompts (identity, context, instruction)
36+ - ` filterRelevantMessages() ` excludes agent's own messages
37+ - ` filterStatusMessages() ` removes Continue CLI status output
38+ - ` isStatusMessage() ` detects and filters loading/initialization messages
39+ - Health check with ` --version ` flag, fallback to ` --help `
40+ - ** Updated** : ` internal/registry/agents.json `
41+ - Added Continue entry with npm install/upgrade/uninstall commands
42+ - Marked ` requires_auth: true `
43+ - Documentation: https://docs.continue.dev/cli
44+ - ** Updated Tests** :
45+ - ` pkg/adapters/adapters_test.go ` - Added TestContinueAgentInitialization
46+ - ` internal/registry/registry_test.go ` - Updated agent count from 15 to 16
47+ - ** Documentation Updates** :
48+ - README.md - Added Continue to supported agents list
49+ - CLAUDE.md - Added Continue CLI technical details
50+
51+ ### Benefits
52+ - ✅ ** TUI and Headless Modes** : Works in both interactive and automated contexts
53+ - 🚀 ** Model Flexibility** : Specify any supported model via ` --model ` flag
54+ - 🔇 ** Clean Output** : ` --silent ` flag removes internal reasoning tags
55+ - 🔐 ** Secure Authentication** : Centralized auth via Continue Mission Control
56+ - 🎯 ** Multi-Agent Ready** : Filters messages for clean conversation participation
57+ - 📝 ** Well Documented** : Complete examples and troubleshooting guides
58+
1059## [ 0.6.0] - 2025-10-26
1160
1261### Added
@@ -1099,7 +1148,8 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
10991148- ** Clean Message Display** : Smart consolidation of headers and proper paragraph formatting
11001149- ** Cost Transparency** : See exactly how much each conversation costs
11011150
1102- [Unreleased]: https://github.com/kevinelliott/agentpipe/compare/v0.2.1...HEAD
1151+ [Unreleased]: https://github.com/kevinelliott/agentpipe/compare/v0.7.0...HEAD
1152+ [0.7.0]: https://github.com/kevinelliott/agentpipe/compare/v0.6.0...v0.7.0
11031153[v0.2.1]: https://github.com/kevinelliott/agentpipe/compare/v0.2.0...v0.2.1
11041154[v0.2.0]: https://github.com/kevinelliott/agentpipe/compare/v0.1.5...v0.2.0
11051155[v0.1.5]: https://github.com/kevinelliott/agentpipe/compare/v0.1.4...v0.1.5
0 commit comments