Need help implementing this on your team?
We'll install the skills, customize them to your workflow, and run the output through your team's combined LinkedIn network. Skills do the research, Linkd does the motion.
Drop a call transcript, a Slack brief, a one-line request, or a rough ICP. The skill extracts targeting signals, maps them to structured Sales Nav filters, and emits a single paste-safe URL plus a readable breakdown. Opinionated about keyword discipline.
URL encoding for Sales Nav is solved. The hard part is fidelity: extracting what the input actually says, knowing when to ask instead of guess, and refusing to fabricate filters the input never established.
The skill accepts every input shape: a meeting transcript, a one-line NL request, a pasted Slack note, a rough ICP doc, or a customer profile. It normalizes each into the same internal signal set, then runs them through the same pipeline. No silent defaults, no padded filters.
Strongest opinion in the skill: the Boolean keywords field is a last resort, not a primary tool. Most "narrow Sales Nav searches" come from keyword stuffing that structured filters could handle better with higher recall. The skill enforces this on every search it builds.
Transcript, brief, NL one-liner, ICP, mixed. Each gets a different reading style, but all normalize into the same internal signals.
Pulls every targeting signal actually present, tied to a quote. No anchor means the category is absent, not filled with a default.
Direct, inferred, ask, unresolvable. Persona or title with no basis in the input is always an ASK. No safe default for who to target.
Structured filters first. Keywords field is a last resort, capped at 2 terms. The skill asks before stuffing concepts into keywords that have a structural alternative.
Paste-safe URL plus a readable breakdown with confidence tags per filter. Seat-gated filters flagged. Session params stripped.
Most narrow Sales Nav searches come from keyword stuffing. The skill defaults to structured filters and caps keywords at 2 terms even when they're justified.
If the input doesn't establish who to target, the skill stops and asks. There is no safe default for persona. "Just build it" instructions don't override this.
Sales Nav renumbers seniority IDs periodically. The old 1 to 10 scale silently drops chips. The skill uses confirmed-live 3-digit IDs and recommends paste-testing.
Spotlights (Posted on LinkedIn, etc.) are Advanced or Advanced Plus only. They silently drop on Core seats. Every search that uses one flags the seat tier explicitly.
The URL is one line, stripped of session params, ready to paste into Sales Nav or feed into Apify or Linked Helper. The breakdown is for your eyes only: confidence tags per filter, seat-gating notes, and an explicit list of what was not applied and why.
The download is a zip containing two things: the .skill file (what you upload to Claude) and a SKILL.md sidecar (the human-readable instructions). Unzip first, then follow the steps for your Claude.
.skill file from the unzipped bundle..skill file too (it's also a zip).~/.claude/skills/web_search and file output enabled.This is exactly what Claude reads when the skill triggers. No hidden prompts, no separate config. What you see here is what you get.
We'll install the skills, customize them to your workflow, and run the output through your team's combined LinkedIn network. Skills do the research, Linkd does the motion.