A command for inspecting or planning changes to Claude's typed workspace-window bindings. A binding records which real session window belongs to a workspace and what it is responsible for.

Use it to inspect window and runtime status, plan an exact replacement, or decommission a window after the required identity and closure checks succeed.

It helps identify the correct window and prevents accidental changes based only on names, titles, terminal panes, or old records.

windows — Command by GxFn

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods

npx agentmods add commands/gxfn/wakeflow/windows

Clone the repo

git clone --depth 1 https://github.com/GxFn/Wakeflow

Per session 12 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 369 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. Scan, not verified.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5	$0.00012	$0.00369
Opus 5	$0.00006	$0.00185
Sonnet 5	$0.00002	$0.00074
Haiku 4.5	$0.00001	$0.00037

Measured yesterday against content hash 16ee092a2cb1, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

windows scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

plugins/claude-code-wakeflow/commands/windows.md · 35 lines

What it actually says

Manage typed Claude window identity without reading or writing private runtime files directly.

Call wakeflow_replace_windows operation=inspect and wakeflow_status operation=inspect. Render stable window ID, role, responsibility, binding status, and redacted runtime status. Never display a raw session handle or locator.
With no mutation request, stop after reporting. A title, tmux pane, cwd, or legacy window-host record is not binding authority.
For replacement, require one exact typed windowId and call wakeflow_replace_windows operation=replace with the owner-specific closed request. Execute only the returned host-neutral intent through the v3 Claude activation owner, then call wakeflow_register_window operation=register with the final real session handle.
For decommission, first obtain the exact v3 close/absence evidence and call operation=decommission with the current binding tuple. Claude closure is machine-verifiable only after exact close and absence probe both succeed.
If the v3 activation/close owner is unavailable, report the blocked intent and stop. Do not fall back to the public-v2 helper, thread-registry, window-host, semantic window name, or a same-named worktree.

Pod first materialization uses wakeflow_pod_open operation=launch-preview/launch-apply; existing materialization uses operation=inspect-materialization. Do not route Pod identity through baseline window replacement or recreate a worktree during inspection.

This command never dispatches task work.

$ARGUMENTS

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

yesterday First seen · 35 lines · 12 tokens per session scan A 16ee092a2cb1

Subscribe to this mod's changes

windows is a command published in the GitHub repository GxFn/Wakeflow (5 stars, last pushed 8d ago), licensed MIT. It adds 12 tokens to every session and 369 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Code review Planning Productivity agent-workflow ai-agents codex-plugin control-plane evidence-review local-first

release-flow

Portable gated release flow — deterministic gate + vendor-agnostic independent reviewer, runnable from any harness.

coo1white/cool-workflow · 20 tokens

release

Prepare a release (step 1 of 2) — version bump and PR. Cutting the tag is a separate operator-only step.

coo1white/cool-workflow · 27 tokens

fieldtheory

Explain and drive the Field Theory CLI — bookmark-sourced seeds, repo-aware possibility runs that score ideas onto a 2x2 grid, and the interconnected .md files they leave behind. Trigger when the user asks about ft, bookmarks, seeds, ft possible (or the older name ft ideas), grids, dots/nodes, or how to turn saved…

afar1/fieldtheory-cli · 88 tokens

windows

Getting it into your agent

What it costs to keep this loaded

Grade A, and why

Nothing flagged

What it actually says

What this file has done since we first saw it

Other commands, from other repositories

release-flow

release

fieldtheory

lians-recall

scorecard

review