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 gaelic-ghost/socket --skill safari-mcp-workflowgit clone --depth 1 https://github.com/gaelic-ghost/socketWrote 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/gaelic-ghost/socket/safari-mcp-workflow)<a href="https://agentmods.dev/skills/gaelic-ghost/socket/safari-mcp-workflow"><img src="https://agentmods.dev/badge/skills/gaelic-ghost/socket/safari-mcp-workflow/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/gaelic-ghost/socket/safari-mcp-workflow"><img src="https://agentmods.dev/badge/skills/gaelic-ghost/socket/safari-mcp-workflow.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 | $0.00048 | $0.01129 |
| Opus 5.5 | $0.00019 | $0.00452 |
| Sonnet 5.5 | $0.00010 | $0.00226 |
| Haiku 4.5 | $0.00005 | $0.00113 |
Grade A, and why
safari-mcp-workflow 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 1mo 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 — 80 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Safari MCP Workflow
Purpose
Use Safari MCP for evidence from a live Safari Technology Preview tab. It owns browser-runtime observation and authorized interaction, not Safari extension architecture, app-to-Safari control, Web Inspector extension design, or cross-browser certification.
When To Use
- Use for Safari-specific rendering, DOM, console, network, screenshot, interaction, accessibility, performance, responsive, or print-media investigation.
- Use when a local web app needs Safari evidence before a fix is proposed or accepted.
- Use
safari-extension-control-workflowfor Safari Web Extension, Safari App Extension, SafariServices, or Web Inspector extension architecture. - Use
xcode-testing-workflowfor XCTest, XCUITest, simulator, or Xcode-run validation. - Use
explore-apple-swift-docsfor Apple documentation lookup that does not need a live tab.
Single-Path Workflow
- Scope the target before connecting: name the origin, tab, requested state, permitted interactions, and any sensitive data boundary. Do not inspect unrelated tabs.
- Preflight Safari Technology Preview 247 or later, its
safaridriverexecutable, and Safari's Enable remote automation and external agents setting (shown as Allow remote automation in some builds). Readreferences/setup-and-privacy.md. - Register Safari MCP only with explicit approval. Use the documented local
safaridriver --mcpcommand; do not bundle a server, silently alter global MCP configuration, or claim a connection before listing its tools. - Establish page state with tab, URL, loading, and
get_page_contentevidence before evaluating JavaScript or taking a screenshot. Use its nodeuidvalues when an interaction must target a specific element. Then collect the smallest useful combination of console, network, DOM, JavaScript, and screenshot evidence. - Perform only the authorized interaction sequence; batch related reversible steps through
page_interactionswith a stated purpose and the relevant nodeuidwhen that makes the recorded sequence clearer. Do not submit forms, purchase, delete, change account state, or reveal secrets without a fresh user confirmation immediately before the effect. - For accessibility, inspect semantic markup, labels, ARIA, keyboard reachability, and relevant visual contrast evidence. Report findings as focused checks, not a compliance certification.
- For performance, use bounded page metrics and network timings. State whether the evidence is a one-run observation, not a benchmark or regression guarantee.
- Re-check the requested outcome after each change with independent evidence. Read
references/evidence-and-validation.mdfor the reporting contract. - Return the target, Safari Technology Preview version, observed evidence, authorized actions, unresolved uncertainty, and one focused handoff.
What ships with it
3 files 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.
- 1mo ago First seen · 80 lines · 48 tokens per session scan A b80e76808bc7
safari-mcp-workflow is a skill published in the GitHub repository gaelic-ghost/socket (7 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 48 tokens to every session and 1,129 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 2026-09-03.
Other skills, from other repositories
bug-fix
Use when a bug, regression, or unexpected behaviour is reported and the user wants it fixed end to end — "why is X broken", "this stopped working after Y", "fix this crash". Also use when a symptom is known but its cause is not. Not for building new behaviour, and not for a change whose cause is already proven.
bug-investigate
Use when someone wants to know WHY something is broken and has not asked for it to be fixed — "why is this happening", "what's causing this error", "find out what's wrong", "diagnose this before we decide". Also use before committing to a fix, when the cause is unknown and the decision depends on it. Not for fixing …
operating-cadence
Designs the rhythm an organization runs on — which reviews happen weekly, monthly and quarterly, what each one decides, who owns the numbers presented, and how a signal at the front line reaches the people who can act on it. Use this to set up a management operating system, fix a meeting calendar that produces no…
serena
Serena code intelligence — LSP-powered symbol navigation, diagnostics, and targeted code surgery. Activate before complex refactors, cross-file analysis, or when graph tools need symbol-level depth.
browser-agent
Autonomous human-like browser control and professional QA/QC testing. Powered by agent-browser (Rust native CDP, zero Playwright) with ISTQB test matrix generation, automated testcase execution, console/network error tracing, and defect reporting.
diagnose
Structured 6-phase debugging. Build feedback loop first, reproduce deterministically, hypothesize with ranked falsifiable theories, instrument one variable at a time, fix with regression test, cleanup. Use when a bug exists, tests fail unexpectedly, or behavior is wrong and cause is unknown.