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.
Live
Discovery
Competitive
Their public customers. With the evidence trail.
Point the skill at a competitor URL. It finds every company publicly identified as a customer, case studies, logo walls, testimonials, with a source link per row. Partner-page logos and aspirational copy get dropped, not silently included.
CD
competitor-customer-discovery.zip
18.0 KB
·
Bundle: .skill file + SKILL.md inside
·
v1.0
Public customers, not pattern-matched guesses.
Customers a competitor names publicly have already (a) bought from a similar vendor, (b) been willing to be named, and (c) demonstrated budget for the category. That's stronger qualification than most prospecting work produces, IF the list is actually accurate.
The skill grades evidence in three tiers. Tier 1: case study pages, logo walls, public testimonials, named G2 reviews. Tier 2: job postings mentioning the tool, engineering blog posts, community references. Tier 3 (dropped unless requested): partner pages, integration listings, "companies in the ICP".
For displacement outreach specifically, where the user is going to claim "I saw you use [competitor]", the skill enforces Tier 1 only. A wrong claim there destroys the conversation. The bar is intentionally high.
Any of these will work.
Who are [Competitor]'s public customers?
Find me the case studies on [vendor.com].
Logos for [Competitor]
Pull a list of companies using [tool] for displacement outreach.
Five steps.
01
Confirm the target
Disambiguate name collisions. "Atlas" could be a dozen products. Then ask whether the list is for lookalikes, displacement, or research, affects the tier bar.
02
Hunt Tier 1 first
Case study pages, logo walls, customer pages on the competitor's own site. Then third-party: G2/Capterra reviewer companies, press releases, conference co-presents.
03
Broaden to Tier 2 if needed
Job postings mentioning the tool. Engineering blog posts. Community references. Marked clearly as Tier 2; never silently promoted to Tier 1.
04
Filter ruthlessly
Drop partner-page logos. Drop dead/acquired companies. Drop case studies older than 24 months without flagging. Resolve subsidiaries vs. parents.
05
Emit JSON + CSV
JSON for the chain (lookalike-expansion, account-list-tierer). CSV for Sales Nav upload. Every row carries evidence tier, source URL, and date.
No invented customers. Ever.
!
Plausibility is not evidence
A real fintech that "would obviously use" the competitor doesn't go on the list without a source. Worst possible failure: outreach claiming use to companies that don't.
@
Partner pages aren't customer pages
A logo on a partner / integration / co-marketing page is categorically different from a customer logo. The skill never confuses them.
#
Aspirational copy gets dropped
"Acme works with leading fintechs" is marketing, not a customer attribution. Surfaced as dropped, not silently included to pad the count.
?
Displacement bar is higher
When the user is going to pitch "we saw you use [competitor]", the skill enforces Tier 1 only. A wrong claim there ends the conversation.
Evidence per row, every row.
Output is JSON for downstream skills (lookalike-expansion, account-list-tierer, account-research-brief) plus a Sales Nav-compatible CSV. Each row carries the evidence tier, the source URL, the date, and notes on anything ambiguous.
Sample row
{
"name": "Globex",
"domain": "globex.io",
"evidence_tier": "TIER_1",
"evidence_type": "case_study",
"evidence_url":
"acme.com/customers/globex",
"evidence_date": "2025-09",
"domain_confidence": "HIGH"
}
Customer contract
- Output
- JSON + CSV
- Evidence tiers
- TIER_1 / TIER_2 / dropped
- Required tools
- web_search, web_fetch
- Domain confidence
- HIGH / MED / LOW flagged
- Sales Nav-ready
- Yes (CSV)
- Pairs with
- lookalike-expansion
Pick your Claude. Drop in the skill.
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.
⌘
Most people
Claude.ai or Desktop
- Download and unzip the bundle.
- Open or create a Project in Claude.
- Go to Project knowledge, click Add content.
- Upload the
.skillfile from the unzipped bundle. - Start a new chat in that project. Ask it to run the skill.
Download bundle ↓
No restart needed
›_
Developers
Claude Code
- Download and unzip the bundle.
- Unzip the inner
.skillfile too (it's also a zip). - Move the resulting folder into
~/.claude/skills/ - Restart Claude Code.
- Type a prompt that fits the trigger. The skill loads automatically.
Download bundle ↓
macOS / Linux / WSL
{ }
Builders
API or Agent SDK
- Download and unzip the bundle.
- Open the SKILL.md sidecar to read the full instructions.
- Load via the Skills API, or paste the instructions into your agent's system prompt.
- Make sure your agent has
web_searchand file output enabled. - Ship it.
View SKILL.md ↓
v1.0
Read the full instructions.
This is exactly what Claude reads when the skill triggers. No hidden prompts, no separate config. What you see here is what you get.
competitor-customer-discovery-SKILL.md
⌄
competitor-customer-discovery-SKILL.md
Loading SKILL.md...