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 dpc00/sublime-mcp --skill screenshot-routinggit clone --depth 1 https://github.com/dpc00/sublime-mcpWrote 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/dpc00/sublime-mcp/screenshot-routing)<a href="https://agentmods.dev/skills/dpc00/sublime-mcp/screenshot-routing"><img src="https://agentmods.dev/badge/skills/dpc00/sublime-mcp/screenshot-routing/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/dpc00/sublime-mcp/screenshot-routing"><img src="https://agentmods.dev/badge/skills/dpc00/sublime-mcp/screenshot-routing.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 | $0.00046 | $0.00694 |
| Opus 5.5 | $0.00018 | $0.00278 |
| Sonnet 5.5 | $0.00009 | $0.00139 |
| Haiku 4.5 | $0.00005 | $0.00069 |
Grade A, and why
screenshot-routing 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 7d 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 — 55 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Screenshots, prompts and window state (sublime-mcp tools)
Scope: the Sublime that hosts this MCP server, reached through sublime-mcp's own tools. Capturing any other process is not sublime-mcp's job and is not covered here.
Which call to use
| what you want | call |
|---|---|
| the editor window, or part of it | ui_screenshot |
| a dialog, menu or popup | ui_screenshot with kind (dialog, menu) |
| the truth about a native window (title, buttons, message text) | list_native_windows |
| text state — sheets, selection, file contents, syntax | get_sheets, get_selection, get_active_file |
ui_screenshot:
- select with
window_title(substring),hwnd, orkind; - always pass
max_width(900 reads a popup fine) andregionwhen only part matters; inline=falsewhen you only need the file rather than the image;- it does not wait on a stuck UI thread:
the window is not answeringmeans the window failed a WM_NULL check and was not asked to render — wait for Sublime to be idle and retry; - a
fallback: "screen"in the result meansPrintWindowmissed its 2.5 s deadline and the picture was copied off the desktop, so anything covering the window is in it.
A prompt on screen means the work is not done
A picker or dialog left up means the package is waiting for an answer; treat the step as unfinished.
- ST4 input-handler pickers are invisible to
window.views(),active_view(), the UIA tree andlist_native_windows— only a screenshot shows them. - Native modals do appear in
list_native_windows, with their text and buttons, and it works while the main thread is blocked. - Answer a modal with
dismiss_native_window {"hwnd":…, "action":"button", "button":"Yes"}. It needs no focus and works while Sublime is blocked; an AXpress()on the same button silently does nothing.
Empty window list = wrong process, not a broken tool
{"error": "no window ...", "windows": []} means the client is talking to a windowless Sublime
process, typically a leftover from before a restart.
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.
- 7d ago First seen · 55 lines · 46 tokens per session scan A fcd839f81e72
screenshot-routing is a skill published in the GitHub repository dpc00/sublime-mcp (4 stars, last pushed 8d ago), licensed MIT. It adds 46 tokens to every session and 694 once invoked, about $0.0002 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 2026-11-09.
screenshot-routing compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| screenshot-routingdpc00/sublime-mcp | Skill | 4 | 8d ago | MIT | A | 46 | Claude Code, Codex |
| report-issue-localcyanheads/obsidian-mcp-server | Skill | 706 | 5d ago | Apache-2.0 | A | 44 | Claude Code |
| opik-diagnosecomet-ml/opik-mcp | Skill | 226 | 2d ago | Apache-2.0 | A | 187 | Claude Code |
| opik-explaincomet-ml/opik-mcp | Skill | 226 | 2d ago | Apache-2.0 | A | 113 | Claude Code |
| lsp-inspectblackwell-systems/agent-lsp | Skill | 180 | today | MIT | A | 126 | Claude Code |
| lsp-refactorblackwell-systems/agent-lsp | Skill | 180 | today | MIT | A | 57 | Claude Code |
| lsp-safe-editblackwell-systems/agent-lsp | Skill | 180 | today | MIT | A | 59 | Claude Code |