Add compact runtime status CLI

[danieldear]

Summary

• add agent007 status for compact local runtime/session visibility
• show active/running/blocked/failed counts plus recent session rows with lifecycle, workflow progress, age, and action hints
• add --json output for tooling/automation

Validation

• cargo fmt --all -- --check
• cargo test -p agent007 runtime_status
• cargo test -p agent007 parse_status_subcommand
• cargo check -p agent007
• cargo run -p agent007 -- status --limit 3
• cargo run -p agent007 -- status --limit 1 --json

[Copilot]

Pull request overview

Adds a top-level agent007 status command to provide compact local runtime/session visibility, including optional JSON output for tooling.

Changes:

• Registers the new status CLI subcommand and dispatch path.
• Adds runtime status snapshot building, JSON output, and compact table rendering.
• Adds parser and formatting helper tests for the new command.

Reviewed changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated 3 comments.

File	Description
crates/cli/src/main.rs	Adds the Status command variant, dispatch handling, and CLI parse test.
crates/cli/src/commands/runtime_status.rs	Implements status snapshot generation, workflow/session summarization, JSON output, and table formatting.
crates/cli/src/commands/mod.rs	Exposes the new runtime_status command module.

Comments suppressed due to low confidence (2)

crates/cli/src/commands/runtime_status.rs:187

• This only emits workflow-specific hints when a step is already running (or blocked/failed). Hosted workflows can also be waiting with no running step but with one or more ready/pending steps whose dependencies are satisfied; the existing runtime API handles that as ready/host should submit ready steps. Those sessions will fall through to monitor run, which is misleading because nothing is actually running and the host needs to submit the next step.

        if !workflow.running_steps.is_empty() {
            return "monitor running steps";
        }

crates/cli/src/commands/runtime_status.rs:249

• Errors are truncated but not normalized like task and output_preview, so a multi-line workflow error can spill out of the compact row format and make the status table hard to parse. Replace embedded newlines before truncating here as well.

            println!("  error: {}", truncate(error, 88));

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.