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 manishiitg/coding-agent-loop --skill work-ui-controlgit clone --depth 1 https://github.com/manishiitg/coding-agent-loopWrote 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/manishiitg/coding-agent-loop/work-ui-control)<a href="https://agentmods.dev/skills/manishiitg/coding-agent-loop/work-ui-control"><img src="https://agentmods.dev/badge/skills/manishiitg/coding-agent-loop/work-ui-control/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/manishiitg/coding-agent-loop/work-ui-control"><img src="https://agentmods.dev/badge/skills/manishiitg/coding-agent-loop/work-ui-control.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 | $0.00037 | $0.00743 |
| Opus 5.5 | $0.00015 | $0.00297 |
| Sonnet 5.5 | $0.00007 | $0.00149 |
| Haiku 4.5 | $0.00004 | $0.00074 |
Grade A, and why
work-ui-control 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 24d ago.
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 — 57 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Crew UI control
Use list_ui_capabilities, get_ui_state, and perform_ui_action only in an
interactive Crew project chat that actually exposes them. They control the
visible right-side project pane; they do not edit project data, run automation,
or prove that the user read the result.
Receipt contract
- Use only view IDs, actions, and targets returned by
list_ui_capabilities. appliedconfirms that the visible panel shell acknowledged the action. It does not prove that every request inside the panel finished successfully.accepted,applying,expired,browser_disconnected, and other failures are not success. Report the bounded result instead of claiming the panel opened.- Reuse an
idempotency_keyonly when retrying the exact same action after an uncertain outcome. Never replay an uncertain action with a new key. - Open the one panel that best supports the current reply. Do not repeatedly override a panel the user selected.
Crew views
These are Crew views, not AgentWorks Workflow views:
| View ID | Crew panel | Open it when |
|---|---|---|
report |
Dashboard | A project Dashboard was created or changed, or the user asks to see its visual project information. |
memory |
Memory | MEMORY.md or project-local reusable skills were reviewed or changed. |
database |
Database | Project-owned managed tables or rows were created, changed, or requested. |
files |
Files | The user should inspect project files. Crew opens the panel; it does not deep-link to an individual file through UI control. |
browser |
Browser | Managed browser work begins or the user asks to watch it. The stream updates without repeated refresh actions. |
costs |
Costs and usage | The user asks about this Crew project's model usage or cost. |
workshop |
Automation | The user should inspect project chats, schedules, triggers, or bots. Use only an advertised section target. |
schedules |
Automation compatibility route | Open the schedules or webhook-trigger section using only an advertised target. Prefer workshop when its target expresses the destination. |
identity |
Identity and project setup | The user should inspect project identity, selected model, secrets, or attached folders. |
mcp |
Integrations | The user should inspect MCP servers, skills, bots, or connected accounts. |
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.
- 24d ago First seen · 57 lines · 37 tokens per session scan A 187da93a5aee
work-ui-control is a skill published in the GitHub repository manishiitg/coding-agent-loop (13 stars, last pushed 19d ago), licensed MIT. It adds 37 tokens to every session and 743 once invoked, about $0.0001 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 2026-09-23.
Other skills, from other repositories
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.
apple-tools
Reach Apple PIM data (Calendar, Contacts, Reminders, Messages, Location, Maps, Weather) from a pi session via iMCP + pi's built-in MCP on macOS. Use when the user asks to read/search their Apple Calendar, Contacts, Reminders, iMessages, current Location, Maps, or Weather. Does NOT cover Apple Mail — iMCP exposes no…
pi-dashboard-bootstrap
Make sure the local pi-dashboard server is running before any pi-dashboard action. Use FIRST whenever the user mentions the pi dashboard, pi sessions or pi agents, or asks to start/open the dashboard; also when a dashboard call fails with "connection refused".
skill
Skill "skill" from wavyrai/tmux-ide, covering tmux-ide — claude code skill, when to use, the agent contract, coordinating with other agents and fleet control from the cli.
kb-setup
Set up, initialize, or configure the local markdown knowledge base for a project. Use when the user says "set up the knowledge base", "init the KB", "configure the markdown KB", "index my docs", "add a doc source to the KB", or wants the project's markdown searchable by kb search. Wraps kb init end-to-end through kb…
kb-search
Search the local markdown knowledge base for facts, prior decisions, definitions, people, error codes, config flags, and how-we-fixed-it notes. ALWAYS search here BEFORE answering a project-specific question from memory, guessing, or asking the user. Use whenever you hit an unknown term, an unfamiliar entity, an error…