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 manykarim/rf-mcp --skill rf-locatorsgit clone --depth 1 https://github.com/manykarim/rf-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/manykarim/rf-mcp/rf-locators)<a href="https://agentmods.dev/skills/manykarim/rf-mcp/rf-locators"><img src="https://agentmods.dev/badge/skills/manykarim/rf-mcp/rf-locators/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/manykarim/rf-mcp/rf-locators"><img src="https://agentmods.dev/badge/skills/manykarim/rf-mcp/rf-locators.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 | $0.00057 | $0.01328 |
| Opus 5.5 | $0.00023 | $0.00531 |
| Sonnet 5.5 | $0.00011 | $0.00266 |
| Haiku 4.5 | $0.00006 | $0.00133 |
Grade A, and why
rf-locators 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 2d 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 — 78 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Picking locators with rf-mcp
Inspect first, never guess
In an rf_repl session with the page or app open:
| Library | Inspect | Then check a candidate |
|---|---|---|
| Browser | Get Aria Snapshot css=body (or a narrower container such as main, header, form) |
Get Element Count <locator> must be 1 |
| SeleniumLibrary | Get Source (large) or Get WebElements css:... |
Get Element Count <locator> |
| AppiumLibrary | Get Source |
Get Matching Xpath Count <xpath> |
| PlatynUI | Query /app:* then a scoped Query |
Query <descriptor> only_first=True |
The ARIA snapshot lists roles and accessible names (button "Add to cart", textbox "Email") - the most stable handles a page has. It shows open shadow content but each iframe only as - iframe: snapshot a frame through it with Get Aria Snapshot id=payframe >>> body.
Stability order (most stable first)
- Dedicated test hooks:
[data-testid="checkout-submit"],id=submit - Role + accessible name:
role=button[name="Place order"],role=textbox[name="Email"] - Form semantics:
[name="email"],labeltext,input[type="email"] - Visible text:
text="Sign in"(exact) /text=Sign in(substring, case-insensitive) - Structural CSS/XPath (
div > div:nth-child(3), generated class names such as.css-1x2y3z,>> nth=2) - last resort; they break when markup changes.
Scope inside a stable container instead of using nth: form[action="/checkout"] >> role=textbox[name="Email"], article:has-text("Headphones") >> role=button[name="Add to cart"].
Put every locator in one place (a *** Variables *** section or the resource keyword that owns the page) so a UI change is a one-line fix.
Shadow DOM and iframes
| Situation | Browser | SeleniumLibrary |
|---|---|---|
| Open shadow root | CSS, text=, role= pierce it: login-card >> role=button[name="Sign in"]; XPath does not |
CSS and XPath do not pierce: ${host}= Get WebElement css=login-card then Evaluate $host.shadow_root.find_element("css selector", "#user") |
| iframe | id=payframe >>> id=card; nested or cross-origin: id=outer >>> id=inner >>> text=Confirm |
Select Frame id=payframe ... Unselect Frame (in [Teardown]) |
Closed shadow root (attachShadow({mode: 'closed'})) |
invisible to selectors, Get Aria Snapshot and host Get Text |
shadow_root fails |
What ships with it
8 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.
- 2d ago First seen · 78 lines · 57 tokens per session scan A 5c78e566342d
rf-locators is a skill published in the GitHub repository manykarim/rf-mcp (118 stars, last pushed yesterday), licensed Apache-2.0. It adds 57 tokens to every session and 1,328 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 2026-11-12.
rf-locators compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| rf-locatorsmanykarim/rf-mcp | Skill | 118 | yesterday | Apache-2.0 | A | 57 | Claude Code, Codex |
| e2e-fixturesxberg-io/xberg | Skill | 9.6k | today | MIT | A | 18 | Claude Code |
| test-electron-appPostHog/posthog | Skill | 41k | today | none | A | 112 | Claude Code |
| behave-skillLambdaTest/agent-skills | Skill | 392 | 9d ago | MIT | A | 66 | Claude Code, Codex |
| playwright-testingalinaqi/maggy | Skill | 712 | 1mo ago | MIT | A | 20 | Claude Code |
| running-mobile-demo-flowsPostHog/posthog | Skill | 41k | today | none | A | 103 | Claude Code, Codex |
| visual-validationalinaqi/maggy | Skill | 712 | 1mo ago | MIT | A | 23 | Claude Code |