Install any skill in seconds. Free to start, no credit card required.
Get Started Free →jakeschincariol/li-reply
Handle the replies under the user's own LinkedIn posts - draft answers to every comment, sorted by which ones are worth answering.
Passed 18 of 23 test cases with the skill, and 5 of 23 without it.
and 18 more in the full report
14 fixed4 passed either way4 still failing1 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 14 test cases the skill fixed. Same model, same prompt, with and without the skill.
Below are comments on my post about Kubernetes migration: - @CloudConsulting: 'We offer zero-downtime K8s migration services for small teams. Visit our site!' - @Alex Rivera: 'What tool did you use for ingress routing?' Draft responses for both comments. For @CloudConsulting, give them a warm 3-sentence thank you reply for stopping by.
| Checked | Without skill | With skill |
|---|---|---|
| The response classifies CloudConsulting's comment as NOISE. | Missed | Passed |
| The reply to Alex Rivera answers his question in public. | Passed | Passed |
The instructions your model reads when the skill loads.
The reply thread under your own post is where reach is actually decided. Every reply is a second engagement event on the post, and the first hour of replies does most of the work. But the value is not equal across comments, so this skill sorts before it writes.
The user pastes the comments, ideally with names and roles. Screenshots are fine. Do not scrape the thread with a browser tool.
Sort every comment into one of five buckets and say the count out loud:
| bucket | what it is | what it gets |
|---|---|---|
| LEAD | someone describing the problem you solve | a real answer + a soft open door |
| SUBSTANCE | adds data, disagrees, extends | the longest reply on the thread |
| PEER | a name worth being seen next to | a reply that gives them something |
| SUPPORT | "great post", 🔥, a tag | a like, and a 3-8 word reply at most |
| NOISE | pitch, spam, bad faith | nothing, or one line and out |
Then write in that order, and stop writing when the value stops.
test cases passed, vs 5 / 23 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.