cm-status

Status

Show an overview of current Claude Code, Pi Coding Agent, and Codex activity.

Instructions

First, gather data by calling these tools in parallel:

1. mcp__plugin_cman_cman__list_plans
2. mcp__plugin_cman_cman__list_sessions with limit=10 and exclude_subagents=true
3. mcp__plugin_cman_cman__list_memory
4. mcp__plugin_cman_cman__list_pi_sessions with limit=10
5. mcp__plugin_cman_cman__list_codex_sessions with limit=10

Then use the results to generate the output.

Scope

• Default: Show only entries matching the current working directory.
• If $ARGUMENTS contains all: Show entries from all projects.

To determine the current project, use the last component of the current working directory. Filter plans, sessions, and memory by matching the project/cwd path.

Output format

## Status

### Active Plans
| # | Plan | Sessions | Project |
|---|------|----------|---------|
| 1 | {title} | {count} | {project} |

### Recent Sessions
| # | Agent | Title | Project | When | CLI Resume |
|---|-------|-------|---------|------|------------|
| 1 | Claude | {title} | {project} | {relative_time} | `cd {cwd} && claude --resume {session_id}` |
| 2 | Pi | {title} | {project} | {relative_time} | `cd {cwd} && pi --session {session_id}` |
| 3 | Codex | {title} | {project} | {relative_time} | `cd {cwd} && codex resume {session_id}` |

### Memory Overview
- {project}: {count} files
  - `{full_path_to_file_1}`
  - `{full_path_to_file_2}`

• Keep it scannable — tables for plans and sessions, bullet list for memory.
• For the "Project" column, extract the last directory component from the cd path.
• If there are no active plans, say "No active plans" instead of an empty table.
• For the CLI Resume column, use the exact resume command shown by the MCP tool.
• For the Title column, rephrase raw prompts into brief work descriptions (e.g., "@README.md" → "README editing", "/remember test" → "Search testing").
• For Memory Overview, always include the full file path (e.g., ~/.claude/projects/-Users-.../memory/filename.md) for each file listed in the data.