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.
npx skills add bks-lab/open-bridge --skill tmuxgit clone --depth 1 https://github.com/bks-lab/open-bridgeWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/bks-lab/open-bridge/tmux)<a href="https://agentmods.dev/skills/bks-lab/open-bridge/tmux"><img src="https://agentmods.dev/badge/skills/bks-lab/open-bridge/tmux/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/bks-lab/open-bridge/tmux"><img src="https://agentmods.dev/badge/skills/bks-lab/open-bridge/tmux.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 5.1 | $0.00074 | $0.01015 |
| Opus 5.5 | $0.00030 | $0.00406 |
| Sonnet 5.5 | $0.00015 | $0.00203 |
| Haiku 4.5 | $0.00007 | $0.00102 |
Grade A, and why
tmux scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured today.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 77 lines — stays where its author put it; the contents beside it link to each section on GitHub.
tmux
A second terminal driver beside skills/cmux. It speaks the
workplace JSON protocol (docs/workplace.md § Writing a driver), so the plan, the
briefing, the inbox and the cockpit work the same in any terminal app that runs
tmux.
Optional. It needs the tmux binary ($BRIDGE_TMUX_BIN, else tmux on
PATH, else Homebrew's path). Without it the driver answers an error, which
scripts/workplace.py reads as "tab view unknown", never as "no tabs open".
Wire it
workplace:
driver: {command: ["python3", "${root}/skills/tmux/scripts/tmux_driver.py"]}
Then python3 scripts/workplace.py open --yes [--here "$TMUX_PANE"], launch,
status, send and adopt steer the tabs, exactly as with cmux.
BRIDGE_TMUX_SOCKET=<name> points the driver at a separate server
(tmux -L <name>); the tests use that so they never touch your own tmux.
Mapping
| Workplace | tmux |
|---|---|
| workspace | session (: and . in a name become _, as tmux does) |
| tab | window, its name is the tab name |
ref / id |
pane id of the window's active pane, %3 |
is_self |
the window holds $TMUX_PANE of the caller |
state |
read from the pane: no agent process below the pane is shell; esc to interrupt is working; a permission or numbered choice prompt is needs-you; else waiting |
sessions |
--resume <uuid>, -r <uuid> or --session-id <uuid> in the agent's process args |
All twelve verbs are served: tabs, sessions, open, launch, send,
rename, control, caps, focus, key, screen, close.
What it guarantees
- It never closes a window or a session on its own.
closekills one pane on request, and refuses the last pane of the server, which would end the server. - It never moves the last window out of a session (tmux destroys an emptied session); such a tab is renamed where it is and the report says so.
openandcontrolreuse a window of the same name in the target session.launchdoes not, like cmux:workplace.pyalready skips an agent that runs, and--againasks for a second tab on purpose.launch --target tabopens in the caller's session (here, else$TMUX_PANE) and refuses when neither is known, never the attached session.- A command over 1000 bytes goes into a private launcher file
(
~/.cache/bridge-tmux/commands/, mode 0600, pruned after 7 days), since tmux carries a command in one message of at most 16 KB. Every command keeps a shell open after the agent exits. - A step tmux refuses is an ERROR line, never a success line. No server running is a true "no tabs".
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- today First seen · 77 lines · 74 tokens per session scan A 1fa49807d4e6
tmux is a skill published in the GitHub repository bks-lab/open-bridge (13 stars, last pushed today), licensed MIT. It adds 74 tokens to every session and 1,015 once invoked, about $0.0003 per session on Opus 5.5. 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 2026-10-16.
Other skills, from other repositories
slack
Use the assigned Slack bot from Slack conversations, Paperclip tasks, and routines to read shared discussions and collaborate.
agentmail
Use your assigned AgentMail inbox to read email tasks, explicitly send or reply, and check delivery. Provided automatically by your inbox assignment.
kungfu-agent-onboarding
Use when a user asks to understand, start, inspect, extend, or safely operate installed Kungfu; verify the installed pack, select one intent route, personalize the explanation, and propose one smallest safe next action.
cog-braindump-capture
Capture raw thoughts with automatic domain classification and vault routing.
context-preservation
State capture and restore across context window compactions. Monitors usage thresholds and serializes quality, task, and spec state for seamless continuation.
cog-knowledge-consolidation
Build structured knowledge frameworks from scattered vault notes with source attribution.