migrateforce/candidate-outreach

Summary

This skill generates personalized, high-converting outreach messages to passive candidates. It researches the candidate's background, finds relevant connection points, and crafts messages that stand out from generic recruiter spam.

Inputs

| Field | Type | Required | Description | |-------|------|----------|-------------| | candidate_profile | object | yes | LinkedIn/GitHub profile data | | job_opportunity | object | yes | Role being recruited for | | company_info | object | yes | Company details and selling points | | outreach_channel | enum | yes | linkedin, email, twitter, github | | tone | enum | no | professional, casual, technical | | sequence_stage | number | no | Which follow-up message (1-4) |

Outputs

| Field | Type | Description | |-------|------|-------------| | subject_line | string | Email/InMail subject | | message_body | text | Personalized outreach message | | personalization_points | array | Specific details referenced | | follow_up_date | date | When to send next message | | ab_variant | string | Which template variant used |

Workflow

1. Research Candidate

From LinkedIn:

• Current role and company
• Previous experience
• Education and skills
• Posts and articles
• Mutual connections
• Recent activity

From GitHub:

• Popular repositories
• Recent contributions
• Languages and technologies
• Open source involvement

2. Find Connection Points

PERSONALIZATION HIERARCHY (best to worst):
1. Mutual connection ("Sarah mentioned you...")
2. Their content ("Loved your post on...")
3. Shared experience ("Fellow [school] alum...")
4. Their work ("Your project X is impressive...")
5. Company news ("Congrats on the Series B...")
6. Role relevance ("Your [skill] expertise...")

3. Craft Message

Structure:

1. Personalized hook (why them specifically)
2. Brief company/role intro (1-2 sentences)
3. What's in it for them
4. Soft call to action
5. Easy opt-out

Length:

• LinkedIn InMail: 100-150 words
• Email: 150-200 words
• Twitter DM: Under 280 characters

4. A/B Variants

Generate 2-3 variants testing:

• Subject lines
• Opening hooks
• Call-to-action phrasing
• Tone (professional vs casual)

5. Sequence Planning

| Stage | Timing | Focus | |-------|--------|-------| | 1 | Initial | Personalized intro | | 2 | +3 days | Value add (article, insight) | | 3 | +7 days | Social proof (team, culture) | | 4 | +14 days | Final follow-up |

Message Templates

Initial Outreach (LinkedIn)

Hi [First Name],

[Personalized hook - reference their work/post/project]

I'm reaching out because we're building [brief description] at [Company],
and your experience with [specific skill/project] caught my attention.

We're looking for a [Role] to [key responsibility]. The team includes
[notable person/background], and we just [recent achievement].

Would you be open to a quick call to learn more? No pressure either way.

[Your name]

P.S. [Optional: mutual connection, relevant link, or conversation starter]

Follow-up #1 (Email)

Subject: Re: [Original subject]

Hi [First Name],

Wanted to follow up on my note from last week.

I came across [relevant article/resource] and thought you might find it
interesting given your work on [their project].

[Company] is [brief value prop]. We're solving [problem] for [customers].

If you're not looking right now, I completely understand. But if you know
anyone who might be interested, I'd appreciate a referral.

Best,
[Your name]

Breakup Message

Subject: Closing the loop

Hi [First Name],

I've reached out a few times about the [Role] opportunity at [Company].
I'll assume the timing isn't right.

If that changes, feel free to reach out. I'd still love to connect.

Wishing you continued success at [Current Company].

[Your name]

Personalization Examples

Generic (Bad): > "I came across your profile and thought you'd be a great fit for our role."

Personalized (Good): > "Your talk at ReactConf on state management was excellent—we're dealing with > similar complexity at scale. Would love to chat about how you'd approach our > real-time collaboration challenges."

Generic (Bad): > "We're a fast-growing startup looking for talented engineers."

Personalized (Good): > "We just closed our Series B to build AI tools for recruiters (yes, ironic > that I'm recruiting you). Your work on the recommendation engine at Spotify > is exactly what we need."

Example Usage

User: Write outreach for Jane Doe, Staff Engineer at Google, for our Senior Backend role

Agent: I've researched Jane's profile and crafted this outreach:

**Personalization points found:**
- Recent blog post on distributed systems
- Speaker at KubeCon 2024
- Open source contributor to Envoy
- Mutual connection: Mike (our CTO)

**LinkedIn InMail:**

Subject: Your KubeCon talk + a question

Hi Jane,

Your KubeCon talk on service mesh performance was one of the best I've seen—
especially the latency optimization section. We're running into similar
challenges at [Company] as we scale to 10M requests/sec.

Mike Chen (I think you worked together at Stripe?) suggested I reach out.
We're building [product] and looking for a Staff Engineer to lead our
infrastructure team.

The role: own our Kubernetes platform, scale distributed systems,
mentor 5 engineers. Comp: $350-450k total.

Open to a 20-min call next week?

[Your name]

**A/B Variant (more casual):**

Subject: Fellow Envoy contributor here 👋

Hi Jane,

Saw your Envoy PRs on connection pooling—nice work on the memory
optimization. We're heavy Envoy users at [Company] and running into
an interesting challenge you might have thoughts on...

[continues]

**Recommended send time:** Tuesday 10am PST
**Follow-up scheduled:** Friday if no response

Integration Points

• Upstream: Sourcing lists, LinkedIn Recruiter, GitHub search
• Downstream: ATS candidate creation, interview-scheduler
• Tracking: Open rates, response rates, conversion to interview