docs(tool-guides): correct teams/agents details and de-duplicate hooks events

Author: andkirby

docs/05-tool-guides/agent-teams.md

@@ -21,20 +21,16 @@ This makes "agents talking to each other" explicit and structured instead of ad
 
 ## Claude Code Agent Teams
 
-Claude Code supports team definitions in:
+Agent teams are an experimental Claude Code feature. Enable them by setting `CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1` in the environment where Claude Code runs.
 
-- `.claude/agents/team/**/*.md`
+Use natural-language delegation in chat to involve teammates for specialized work.
 
-The tooling includes:
+Important clarification:
 
-- `/agents` to manage agents
-- `/agents-team` to create and manage agent teams
+- there is no separate `.claude/agents/team/` team-manifest path
+- there is no `/agents-team` command
 
-Team packs can define:
-
-- a main architecture/orchestrator agent
-- specialist teammate agents
-- collaboration patterns between teammates
+Use settings and prompts to shape collaboration behavior, and pair teams with custom agents when you need reusable specialist roles.
 
 ## Team Design Pattern
 

docs/05-tool-guides/hooks.md

@@ -29,25 +29,21 @@ Use hooks for:
 
 Claude Code has first-class lifecycle hooks. In practice, this is the most complete built-in hook system across current coding-agent tools.
 
-### Supported lifecycle events
+### Lifecycle Coverage (High Level)
 
-- `PreToolUse` — before tool execution
-- `PostToolUse` — after tool execution
-- `Notification` — when Claude emits notifications
-- `UserPromptSubmit` — when the user submits a prompt
-- `Stop` — when Claude finishes responding
-- `SubagentStop` — when a subagent finishes
-- `PreCompact` — before context compaction
+Claude hooks can run at key lifecycle points, including:
 
-### Hook handler types
+- session boundaries
+- prompt submission
+- before/after tool calls
+- permission and failure handling
+- sub-agent or teammate lifecycle events
 
-Claude supports multiple hook handler types:
+For the exact and current event names, use the official hooks reference in the References section below.
 
-- **Command hooks** — run shell commands/scripts
-- **Prompt hooks** — inject natural-language guidance at hook time
-- **Agent hooks** — invoke a subagent to evaluate or gate behavior
+### Hook handler types
 
-This makes hooks useful for both deterministic checks (scripts) and policy judgment flows (prompt/agent).
+Claude hooks run command handlers (shell commands/scripts). Handlers can emit stderr/stdout and structured JSON to control blocking and messaging behavior.
 
 ### Configuration locations
 

docs/05-tool-guides/sub-agents.md

@@ -36,8 +36,8 @@ Helpful capabilities:
 
 OpenCode also supports custom agents, with common locations:
 
-- project: `.opencode/agent/*.md`
-- global: `~/.config/opencode/agent/*.md`
+- project: `.opencode/agents/`
+- global: `~/.config/opencode/agents/`
 
 This gives a similar delegation model: keep the main chat as coordinator, dispatch substantial tasks to specialized agents, then merge results.