Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →okhp3/okhp3-linkedin-voice
Apply the OKHP3 brand voice to any LinkedIn-bound text. Use as the FINAL pass on any LinkedIn post or article draft, whether produced by okhp3-linkedin-post or written by hand. Checks for em dashes, paragraph consolidation, ROY-principle density, ending strength, and link-routing rules. This is a filter applied to existing text, not a generator — if there's no draft yet, use okhp3-linkedin-post first.
Passed 17 of 22 test cases with the skill, and 13 of 22 without it.
and 6 more in the full report
5 fixed12 passed either way4 still failing1 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 5 test cases the skill fixed. Same model, same prompt, with and without the skill.
Refine this personal learning update draft: 'Spent the weekend studying Rust concurrency patterns and memory safety guarantees.' Provide the polished text and change log.
| Checked | Without skill | With skill |
|---|---|---|
| The output contains no added company names, corporate handles, or employer branding. | Missed | Passed |
The instructions your model reads when the skill loads.
Portable, identity-agnostic Agent Skill.
This package and its fixtures contain no execution account, person, project, campaign, topic, bias, opinion, or private URL. Use placeholders in public examples and obtain all such context from the current run only.
The platform voice contract. Runs last, on text that already exists.
This package is a final LinkedIn prose filter. It does not generate a new post, publish content, or silently apply LinkedIn formatting to another platform.
references/voice-rules.md.okhp3-linkedin-post should call this skill as its final step, not skip it assuming the draft is already compliant. okhp3-linkedin-angles does not need this skill — angle candidates aren't prose yet.
This skill governs LinkedIn output specifically. Platform-conditional formatting rules apply only when the target surface is known. See references/voice-rules.md for the portable rules.
Return the revised text, then a short change log listing only the rule categories changed or judged. Do not invent new facts, examples, links, hashtags, or employer context while polishing. Preserve any user-specified hashtag count and URL text. If the input is not LinkedIn-bound prose, say so and route to the appropriate skill instead of silently rewriting it.
test cases passed, vs 13 / 22 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.