capabilities

name	capabilities
description	Show what this AgentLite instance can do — installed skills, available tools, and system info. Read-only. Use when the user asks what the bot can do, what's installed, or runs /capabilities.

/capabilities — System Capabilities Report

Generate a structured read-only report of what this AgentLite instance can do.

Main-channel check: Only the main channel has /workspace/project mounted. Run:

test -d /workspace/project && echo "MAIN" || echo "NOT_MAIN"

If NOT_MAIN, respond with:

This command is available in your main chat only. Send /capabilities there to see what I can do.

Then stop — do not generate the report.

How to gather the information

Run these commands and compile the results into the report format below.

1. Installed skills

List skill directories available to you:

ls -1 /workspace/skills/ 2>/dev/null || echo "No skills found"

Each directory is an installed skill. The directory name is the skill name (e.g., agent-browser → /agent-browser).

2. Available tools

Tool names differ by backend. Report the tool families that are actually available in this session:

• Workspace: shell commands, file reads/writes, search, and edits
• Web: web/browser tools when installed by the active backend
• Orchestration: delegation or task tools when available
• MCP: mcp__agentlite__* (messaging, tasks, group management)

3. MCP server tools

The AgentLite MCP server exposes these tools (via mcp__agentlite__* prefix):

• send_message — send a message to the user/group immediately (optional sender for identity labeling)
• schedule_task — schedule a recurring or one-time task (cron / interval / once; group or isolated context)
• list_tasks — list scheduled tasks (main sees all, other groups see their own)
• pause_task — pause a scheduled task
• resume_task — resume a paused task
• cancel_task — cancel and delete a task
• update_task — update an existing task's prompt or schedule
• register_group — register a new chat/group (main only)
• search_actions — discover custom host-side actions (ToolSearch-style query; use before call_action)
• call_action — invoke a host-side action by name with a payload (synchronous; blocks until host returns)

Host-side actions (reached via search_actions + call_action) include the acp_* family when ACP peers are configured. Run search_actions with a relevant query to list what's live on this host.

4. Container skills (Bash tools)

Check for executable tools in the container:

which agent-browser 2>/dev/null && echo "agent-browser: available" || echo "agent-browser: not found"

5. Group info

ls /workspace/group/CLAUDE.md 2>/dev/null && echo "Group memory: yes" || echo "Group memory: no"
ls /workspace/extra/ 2>/dev/null && echo "Extra mounts: $(ls /workspace/extra/ 2>/dev/null | wc -l | tr -d ' ')" || echo "Extra mounts: none"

Report format

Present the report as a clean, readable message. Example:

📋 *AgentLite Capabilities*

*Installed Skills:*
• /agent-browser — Browse the web, fill forms, extract data
• /capabilities — This report
(list all found skills)

*Tools:*
• Workspace: shell, files, search, edits
• Web: available/not installed
• Orchestration: available/not installed
• MCP: send_message, schedule_task, list_tasks, pause/resume/cancel/update_task, register_group, search_actions, call_action

*Container Tools:*
• agent-browser: ✓

*System:*
• Group memory: yes/no
• Extra mounts: N directories
• Main channel: yes

Adapt the output based on what you actually find — don't list things that aren't installed.

See also: /status for a quick health check of session, workspace, and tasks.