agentsys 5.2.1 → 5.3.1

package/.kiro/steering/web-ctl.md

@@ -0,0 +1,106 @@
+---
+inclusion: manual
+name: "web-ctl"
+description: "Use when user asks to \"open a webpage\", \"browse website\", \"log into site\", \"scrape page\", \"interact with web\", \"web automation\", \"browser control\"."
+---
+
+# /web-ctl Command
+
+Browser automation for AI agents. Navigate websites, authenticate with human handoff, and interact with web pages.
+
+## Intent Routing
+
+Parse the user's request and route appropriately:
+
+### Simple Actions (Direct Skill Invocation)
+
+For single-step requests like "go to example.com" or "take a screenshot":
+
+1. Invoke the **web-browse** skill with session name and action
+2. Sessions auto-create on first run command if they don't exist
+
+### Auth Requests
+
+For "log into X" or "authenticate to Y":
+
+1. Create a session named after the service
+2. Invoke the **web-auth** skill
+
+### Multi-Step Flows
+
+For complex requests like "find information on a website" or "fill out a form":
+
+1. Delegate to the **web-session** agent via Task tool
+
+## Quick Reference
+
+```bash
+# Session lifecycle
+node /Users/avifen/.agentsys/plugins/web-ctl/scripts/web-ctl.js session start <name>
+node /Users/avifen/.agentsys/plugins/web-ctl/scripts/web-ctl.js session auth <name> --provider <provider>
+node /Users/avifen/.agentsys/plugins/web-ctl/scripts/web-ctl.js session auth <name> --url <url> [--verify-url <url>] [--verify-selector <sel>]
+node /Users/avifen/.agentsys/plugins/web-ctl/scripts/web-ctl.js session auth <name> --provider <slug> --providers-file <path>
+node /Users/avifen/.agentsys/plugins/web-ctl/scripts/web-ctl.js session providers
+node /Users/avifen/.agentsys/plugins/web-ctl/scripts/web-ctl.js session list
+node /Users/avifen/.agentsys/plugins/web-ctl/scripts/web-ctl.js session end <name>
+node /Users/avifen/.agentsys/plugins/web-ctl/scripts/web-ctl.js session verify <name> --url <url>
+
+# Browser actions
+node /Users/avifen/.agentsys/plugins/web-ctl/scripts/web-ctl.js run <session> goto <url> [--ensure-auth]
+node /Users/avifen/.agentsys/plugins/web-ctl/scripts/web-ctl.js run <session> snapshot
+node /Users/avifen/.agentsys/plugins/web-ctl/scripts/web-ctl.js run <session> click <selector>
+node /Users/avifen/.agentsys/plugins/web-ctl/scripts/web-ctl.js run <session> read <selector>
+node /Users/avifen/.agentsys/plugins/web-ctl/scripts/web-ctl.js run <session> screenshot
+
+# Snapshot control (apply to any action with snapshot output)
+node /Users/avifen/.agentsys/plugins/web-ctl/scripts/web-ctl.js run <session> snapshot --snapshot-depth 3
+node /Users/avifen/.agentsys/plugins/web-ctl/scripts/web-ctl.js run <session> goto <url> --snapshot-selector "css=nav"
+node /Users/avifen/.agentsys/plugins/web-ctl/scripts/web-ctl.js run <session> click <sel> --no-snapshot
+node /Users/avifen/.agentsys/plugins/web-ctl/scripts/web-ctl.js run <session> snapshot --snapshot-collapse
+node /Users/avifen/.agentsys/plugins/web-ctl/scripts/web-ctl.js run <session> snapshot --snapshot-compact
+node /Users/avifen/.agentsys/plugins/web-ctl/scripts/web-ctl.js run <session> snapshot --snapshot-text-only --snapshot-max-lines 50
+node /Users/avifen/.agentsys/plugins/web-ctl/scripts/web-ctl.js run <session> goto <url> --snapshot-full
+node /Users/avifen/.agentsys/plugins/web-ctl/scripts/web-ctl.js run <session> goto <url> --snapshot-collapse --snapshot-depth 4
+
+# Macros
+node /Users/avifen/.agentsys/plugins/web-ctl/scripts/web-ctl.js run <session> select-option <sel> <text>
+node /Users/avifen/.agentsys/plugins/web-ctl/scripts/web-ctl.js run <session> tab-switch <name>
+node /Users/avifen/.agentsys/plugins/web-ctl/scripts/web-ctl.js run <session> modal-dismiss [--accept]
+node /Users/avifen/.agentsys/plugins/web-ctl/scripts/web-ctl.js run <session> form-fill --fields '<json>' [--submit]
+node /Users/avifen/.agentsys/plugins/web-ctl/scripts/web-ctl.js run <session> search-select <sel> <query> --pick <text>
+node /Users/avifen/.agentsys/plugins/web-ctl/scripts/web-ctl.js run <session> date-pick <sel> --date <YYYY-MM-DD>
+node /Users/avifen/.agentsys/plugins/web-ctl/scripts/web-ctl.js run <session> file-upload <sel> <path>
+node /Users/avifen/.agentsys/plugins/web-ctl/scripts/web-ctl.js run <session> hover-reveal <sel> --click <target>
+node /Users/avifen/.agentsys/plugins/web-ctl/scripts/web-ctl.js run <session> scroll-to <sel>
+node /Users/avifen/.agentsys/plugins/web-ctl/scripts/web-ctl.js run <session> wait-toast [--dismiss]
+node /Users/avifen/.agentsys/plugins/web-ctl/scripts/web-ctl.js run <session> iframe-action <iframe> <action> [args]
+node /Users/avifen/.agentsys/plugins/web-ctl/scripts/web-ctl.js run <session> login --user <u> --pass <p>
+node /Users/avifen/.agentsys/plugins/web-ctl/scripts/web-ctl.js run <session> next-page
+node /Users/avifen/.agentsys/plugins/web-ctl/scripts/web-ctl.js run <session> paginate --selector <sel> [--max-pages N] [--max-items N]
+node /Users/avifen/.agentsys/plugins/web-ctl/scripts/web-ctl.js run <session> extract --selector <sel> [--fields f1,f2,...] [--max-items N] [--max-field-length N]
+node /Users/avifen/.agentsys/plugins/web-ctl/scripts/web-ctl.js run <session> extract --auto [--max-items N] [--max-field-length N]
+```
+
+## Examples
+
+User: "Open github.com and show me my profile"
+1. Start session "github"
+2. Auth if needed (web-auth skill)
+3. Navigate to github.com/profile
+4. Snapshot and report
+
+User: "Take a screenshot of example.com"
+1. Start session "quick"
+2. Goto example.com
+3. Screenshot
+4. End session
+
+User: "Log into Twitter"
+1. Start session "twitter"
+2. Auth with `--provider twitter` (auto-configures URL and success detection)
+3. Report success/failure
+
+User: "Log into GitHub"
+1. Start session "github"
+2. Auth with `--provider github`
+3. Report success/failure

package/AGENTS.md

@@ -1,6 +1,6 @@
 # Project Memory: AgentSys
 
-> **Cross-tool compatible**: This file works as CLAUDE.md (Claude Code) and can be copied to AGENTS.md for OpenCode, Codex, and other AI tools.
+> **Cross-tool compatible**: Works across all 5 platforms - Claude Code, OpenCode, Codex CLI, Cursor, and Kiro.
 
 <project-memory>
 
@@ -150,7 +150,7 @@ agentsys                # Run installer
 <agents>
 ## Agents
 
-39 agents across 9 plugins. Key agents by model:
+43 agents across 14 plugins. Key agents by model:
 
 | Model | Agents | Use Case |
 |-------|--------|----------|
@@ -190,6 +190,8 @@ Platform-aware state directory:
 - Claude Code: `.claude/`
 - OpenCode: `config/.opencode/`
 - Codex: `.codex/`
+- Cursor: `.cursor/`
+- Kiro: `.kiro/`
 </state-files>
 
 <workflow-agents>
@@ -240,7 +242,7 @@ Cannot skip in /next-task:
 - Use CHANGELOG.md for completion tracking (not summary files)
 - BEFORE starting → Read the relevant checklist (`checklists/*.md`)
 - BEFORE delivering any work, especially releases → Go through that checklist item by item
-- 4 platforms: Claude Code + OpenCode + Codex + Cursor - ALL must work
+- 5 platforms: Claude Code + OpenCode + Codex + Cursor + Kiro - ALL must work
 - Agent/Skill pattern: Agents invoke skills, skills have implementation
 - Create PRs for non-trivial changes
 </end-reminder>

package/CHANGELOG.md

@@ -7,7 +7,31 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
-## [Unreleased]
+## [5.3.1] - 2026-03-02
+
+### Fixed
+
+- **Code block Task() transform for Kiro** - Phase 9 reviewer Task() calls inside fenced JavaScript code blocks were not being transformed. Fixed with multiline-anchored fence regex that correctly handles backtick template literals inside code blocks.
+
+## [5.3.0] - 2026-03-02
+
+### Added
+
+- **Kiro platform support (#276, #278)** - agentsys now installs to Kiro as a 5th platform alongside Claude Code, OpenCode, Codex CLI, and Cursor. Use `agentsys --tool kiro` or `agentsys install <plugin> --tool kiro` to install. Commands become steering files in `.kiro/steering/` (with `inclusion: manual` frontmatter), skills are copied to `.kiro/skills/` (standard SKILL.md format), and agents are converted to JSON in `.kiro/agents/`. All content is project-scoped under `.kiro/`. Platform detection uses `.kiro/` directory presence.
+
+- **Kiro subagent transforms (#279, #280)** - Task() calls transform to `Delegate to the \`agent\` subagent` with prompt context. AskUserQuestion transforms to markdown numbered-list prompts. Plugin namespace prefixes are stripped.
+
+- **Kiro parallel agent adaptation** - Workflows spawning 4+ parallel reviewer agents (next-task Phase 9, audit-project Phase 2) are automatically adapted for Kiro's experimental 4-agent limit. `installForKiro()` generates two combined reviewer agents (`reviewer-quality-security`, `reviewer-perf-test`). `transformCommandForKiro()` detects consecutive reviewer delegations and rewrites as try-4-then-fallback-to-2 pattern.
+
+- **Subagent comparison documentation** - CROSS_PLATFORM.md now has a platform comparison table for subagent capabilities (spawning, parallelism, teams, ACP) across all 5 platforms.
+
+### Changed
+
+- **`transformAgentForKiro()` refactored** - Now reuses `discovery.parseFrontmatter()` instead of custom parsing. Supports YAML array syntax for tools field.
+
+### Fixed
+
+- **Copy-paste bug in `installForCursor()`** - Skill transform was calling `transformSkillForKiro` instead of `transformSkillForCursor`.
 
 ## [5.2.1] - 2026-03-01
 

package/README.md

@@ -21,7 +21,7 @@
 </p>
 
 <p align="center">
-  <b>14 plugins · 43 agents · 30 skills (across all repos) · 26k lines of lib code · 3,357 tests · 4 platforms</b><br>
+  <b>14 plugins · 43 agents · 30 skills (across all repos) · 26k lines of lib code · 3,357 tests · 5 platforms</b><br>
   <em>Plugins distributed as standalone repos under <a href="https://github.com/agent-sh">agent-sh</a> org — agentsys is the marketplace &amp; installer</em>
 </p>
 
@@ -30,7 +30,7 @@
 </p>
 
 <p align="center">
-  <b>Built for Claude Code · Codex CLI · OpenCode · Cursor</b>
+  <b>Built for Claude Code · Codex CLI · OpenCode · Cursor · Kiro</b>
 </p>
 
 <p align="center"><em>New skills, agents, and integrations ship constantly. Follow for real-time updates:</em></p>
@@ -51,7 +51,7 @@ An agent orchestration system — 14 plugins, 43 agents, and 30 skills that comp
 
 Each agent has a single responsibility, a specific model assignment, and defined inputs/outputs. Pipelines enforce phase gates so agents can't skip steps. State persists across sessions so work survives interruptions.
 
-The system runs on Claude Code, OpenCode, Codex CLI, and Cursor. Install via the marketplace or the npm installer, and the plugins are fetched automatically from their repos.
+The system runs on Claude Code, OpenCode, Codex CLI, Cursor, and Kiro. Install via the marketplace or the npm installer, and the plugins are fetched automatically from their repos.
 
 ---
 
@@ -916,12 +916,13 @@ Every command works standalone. [`/deslop`](#deslop) cleans code without needing
 npm install -g agentsys && agentsys
 ```
 
-Interactive installer for Claude Code, OpenCode, Codex CLI, and Cursor.
+Interactive installer for Claude Code, OpenCode, Codex CLI, Cursor, and Kiro.
 
 ```bash
 # Non-interactive install
 agentsys --tool claude              # Single tool
 agentsys --tool cursor              # Cursor (project-scoped skills + commands)
+agentsys --tool kiro                # Kiro (project-scoped steering + skills + agents)
 agentsys --tools "claude,opencode"  # Multiple tools
 agentsys --development              # Dev mode (bypasses marketplace)
 ```
@@ -972,7 +973,7 @@ The system is built on research, not guesswork.
 - 1,818 tests passing
 - Drift-detect validated on 1,000+ repositories
 - E2E workflow testing across all commands
-- Cross-platform validation (Claude Code, OpenCode, Codex CLI, Cursor)
+- Cross-platform validation (Claude Code, OpenCode, Codex CLI, Cursor, Kiro)
 
 **Methodology:**
 - `/perf` investigation phases based on recorded real performance investigation sessions

package/agent-knowledge/AGENTS.md

@@ -13,6 +13,7 @@
 | GitHub Org Project Management (Multi-Repo OSS) | github-org-project-management.md | 15 | deep | 2026-02-21 |
 | GitHub Organization Structure Patterns | github-org-structure-patterns.md | 18 | deep | 2026-02-21 |
 | Multi-Product Org Docs & Website Architecture | multi-product-org-docs.md | 42 | deep | 2026-02-21 |
+| ACP with Codex, Gemini, Copilot, Claude | acp-with-codex-gemini-copilot-claude.md | 24 | medium | 2026-03-02 |
 
 ## Trigger Phrases
 
@@ -36,8 +37,6 @@ Use this knowledge when user asks about:
 - "Codex MCP" -> ai-cli-advanced-integration-patterns.md
 - "Crush OpenCode" -> ai-cli-advanced-integration-patterns.md
 - "Agent SDK hosting" -> ai-cli-advanced-integration-patterns.md
-- "A2A protocol" -> ai-cli-advanced-integration-patterns.md
-- "Cross-tool MCP" -> ai-cli-advanced-integration-patterns.md
 - "AI CLI server mode" -> ai-cli-advanced-integration-patterns.md
 - "Claude Code as service" -> ai-cli-advanced-integration-patterns.md
 - "Plugin distribution patterns" -> skill-plugin-distribution-patterns.md
@@ -117,6 +116,27 @@ Use this knowledge when user asks about:
 - "Diataxis framework" -> multi-product-org-docs.md
 - "Pagefind search" -> multi-product-org-docs.md
 - "docs-as-code" -> multi-product-org-docs.md
+- "ACP protocol" -> acp-with-codex-gemini-copilot-claude.md
+- "Agent Communication Protocol" -> acp-with-codex-gemini-copilot-claude.md
+- "A2A protocol" -> acp-with-codex-gemini-copilot-claude.md
+- "agent to agent communication" -> acp-with-codex-gemini-copilot-claude.md
+- "cross AI tool communication" -> acp-with-codex-gemini-copilot-claude.md
+- "MCP vs ACP vs A2A" -> acp-with-codex-gemini-copilot-claude.md
+- "Codex CLI agent protocol" -> acp-with-codex-gemini-copilot-claude.md
+- "Gemini CLI agent protocol" -> acp-with-codex-gemini-copilot-claude.md
+- "Copilot agent communication" -> acp-with-codex-gemini-copilot-claude.md
+- "Claude Code agent teams" -> acp-with-codex-gemini-copilot-claude.md
+- "Claude Agent SDK multi-agent" -> acp-with-codex-gemini-copilot-claude.md
+- "sequential thinking MCP" -> acp-with-codex-gemini-copilot-claude.md
+- "MCP bridge server" -> acp-with-codex-gemini-copilot-claude.md
+- "BeeAI Agent Stack" -> acp-with-codex-gemini-copilot-claude.md
+- "Kiro ACP" -> acp-with-codex-gemini-copilot-claude.md
+- "Cross-tool MCP bridges" -> acp-with-codex-gemini-copilot-claude.md
+- "ACP SDK" -> acp-with-codex-gemini-copilot-claude.md
+- "A2A SDK" -> acp-with-codex-gemini-copilot-claude.md
+- "agent interoperability" -> acp-with-codex-gemini-copilot-claude.md
+- "Agent Cards A2A" -> acp-with-codex-gemini-copilot-claude.md
+- "claude mcp serve cross-tool" -> acp-with-codex-gemini-copilot-claude.md
 
 ## Quick Lookup
 
@@ -192,6 +212,16 @@ Use this knowledge when user asks about:
 | GitBook, ReadMe, Nextra | multi-product-org-docs.md |
 | GitHub Pages docs hosting | multi-product-org-docs.md |
 | Cloudflare Pages, Vercel hosting | multi-product-org-docs.md |
+| ACP, Agent Communication Protocol | acp-with-codex-gemini-copilot-claude.md |
+| A2A, Agent2Agent, agent-to-agent | acp-with-codex-gemini-copilot-claude.md |
+| ACP vs MCP vs A2A | acp-with-codex-gemini-copilot-claude.md |
+| BeeAI, Agent Stack, agent interop | acp-with-codex-gemini-copilot-claude.md |
+| agent teams, subagents, multi-agent | acp-with-codex-gemini-copilot-claude.md |
+| MCP bridge, Gemini bridge, OpenAI bridge | acp-with-codex-gemini-copilot-claude.md |
+| sequential thinking, cross-AI reasoning | acp-with-codex-gemini-copilot-claude.md |
+| Agent Cards, agent discovery | acp-with-codex-gemini-copilot-claude.md |
+| claude mcp serve, claude as MCP server | acp-with-codex-gemini-copilot-claude.md |
+| acp-sdk, a2a-sdk | acp-with-codex-gemini-copilot-claude.md |
 
 ## How to Use