Lean body with no over-explanation of concepts Claude already knows; every section (reports list, when-to-use, invocation, red flags) earns its tokens; not a 2 because there is no unnecessary padding to tighten.

Names the exact tool and parameters — "Call the `check_posthog_loading` MCP tool with: `urls` (required), `posthog_host` (default `https://us.i.posthog.com`)" — plus concrete red-flag remediation, which is executable guidance; not a 2 because details are complete rather than pseudocode-like.

As a simple single-task skill under 50 lines, the single action (invoke the tool then interpret) is unambiguous, and the "Interpreting the result" section supplies clear post-execution checkpoints; not capped at 2 because the operation is read-only, not destructive/batch-mutating.

Under 50 lines with no external references needed, organized into well-signaled sections (When to use / How to invoke / Interpreting the result); not a 2 because the structure is clean and no inline content needs splitting out.

Lists multiple concrete actions — "confirm consistent installation across pages, find pages missing the snippet, detect mismatched API keys or hosts between pages, and verify the load method" — matching the anchor for naming several specific actions; not a 2 because coverage is comprehensive rather than partial.

Answers both what ("Inspect how the PostHog JavaScript SDK is loaded across a list of URLs") and when ("Use to confirm..., find..., detect..., verify..."); the "Use to" clause is equivalent explicit trigger guidance, so it is not capped at 2.

Uses natural domain terms a user would actually say — "PostHog", "snippet", "API keys", "pages", "install" — giving good coverage; not a 2 because it spans the common variations rather than a single keyword.

Narrows to a clearly distinct PostHog-loading niche with specific triggers unlikely to conflict with other skills; not a 2 because the scope is sharply bounded rather than overlapping.