Resources / Signal-Based DM Trigger
Live Outreach Signals

A real reason to reach out. Anchored, not invented.

Give the skill a target prospect or company and your pitch. It hunts for a specific recent signal, a post, a hire, a launch, a funding round, then drafts outreach that references it. Every signal traces to a source URL.

Download skill Install instructions
v1.0 Web search required Source-URL anchored Multiple message formats
SD
signal-based-dm-trigger.zip
16.4 KB · Bundle: .skill file + SKILL.md inside · v1.0
Download skill
Want to read what it does first? View source ↓
What it does

A signal alone isn't outreach.

The highest-converting outreach in cold prospecting references something specific and recent that the prospect would notice you noticed. The principle is simple. The failure modes are not, fake signals, stale signals, trivial signals, and misattributed signals all read worse than no reference at all.

The skill hunts in priority order: personal signals first (a post they wrote, a podcast they appeared on, a job change), then company signals attributed to the prospect, then generic company signals, then time-bound external signals. It stops when it finds 2–3 strong ones rather than exhausting the budget.

Each signal is checked for tone before being used. Layoffs, security breaches, and unclear exec departures get flagged and skipped, they're fresh news, but never cheerful conversation starters. The skill won't open with them.

Try saying

Any of these will work.

Find me a reason to reach out to [LinkedIn URL]
Signal-based DM for this prospect.
What's a timely angle for [Company]?
Draft a connection request that references something real.
How it works

Five steps.

01

Confirm the target

Person or company. Disambiguate before searching, "Apex" could be six companies. Ask the pitch context so signal-relevance can be judged.

02

Hunt in priority order

Personal posts first, then company-attributed signals, then generic company news. Default window: 90 days. Older signals get flagged as a stretch.

03

Verify and score

Every signal gets a source URL and a date. Strength rating HIGH / MEDIUM / LOW. Tone-sensitive signals (layoffs, breaches) get flagged.

04

Pick the angle

A signal isn't outreach until it connects to the pitch. If the signal is "they posted about their dog", that's a signal, but not a useful one for selling sales software.

05

Draft to format

Connection request (<300 chars), DM, InMail, or email. Anchored signal in line 1. No "I hope this finds you well" preambles.

Why it doesn't lie to you

Source-anchored on both ends.

!

Every signal traces to a URL

No "I saw somewhere that...", if it can't be linked, it can't be claimed. Every signal carries a source URL and a date.

@

Tone-sensitive signals get flagged

Layoffs, breaches, suspicious exec departures, surfaced as flags, never used as openers. A "congrats" message on a layoff is a disaster.

#

Weak signals get dropped

A short, real reference beats a polished reference to nothing. If 5–10 searches turn up no HIGH or MEDIUM signal, the skill says so plainly.

?

Character counts respected

A 400-char "connection request" doesn't fit LinkedIn's field. Drafts respect the format limits or note the overflow explicitly.

The output

Signal + draft, paired.

Each signal ships with its source URL, date, strength rating, and the connection to the pitch. Then a draft in the requested format, connection request, DM, InMail, or email, with character counts.

Sample output
SIGNAL #1, HIGH Source: linkedin.com/in/sarahc/ Date: 2026-04-22 (4w ago) What: Post on pricing tiers Angle: Your post on bundling made me think about... DRAFT (DM, 78 words): "Saw your post last month arguing for usage-based..."
Signal contract
Output
Markdown + draft messages
Signals per target
2–3 (high-strength)
Required tools
web_search, web_fetch
Time window
90 days (default)
Formats supported
Connect / DM / InMail / Email
Pairs with
profile-personalization-extractor
Install

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

  1. Download and unzip the bundle.
  2. Open or create a Project in Claude.
  3. Go to Project knowledge, click Add content.
  4. Upload the .skill file from the unzipped bundle.
  5. Start a new chat in that project. Ask it to run the skill.
Download bundle ↓ No restart needed
›_
Developers

Claude Code

  1. Download and unzip the bundle.
  2. Unzip the inner .skill file too (it's also a zip).
  3. Move the resulting folder into ~/.claude/skills/
  4. Restart Claude Code.
  5. Type a prompt that fits the trigger. The skill loads automatically.
Download bundle ↓ macOS / Linux / WSL
{ }
Builders

API or Agent SDK

  1. Download and unzip the bundle.
  2. Open the SKILL.md sidecar to read the full instructions.
  3. Load via the Skills API, or paste the instructions into your agent's system prompt.
  4. Make sure your agent has web_search and file output enabled.
  5. Ship it.
View SKILL.md ↓ v1.0
Source

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.

MD
signal-based-dm-trigger-SKILL.md 11.2 KB · the full skill instructions
⌄
Loading SKILL.md...

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.

Book a demo