icebreaker — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited icebreaker (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
Most first-touch outreach loses on touch 1. The opener has two jobs: prove you know who they are (why-this-person), and prove there's a reason to talk NOW (why-now). Most openers fail both - they're generic on the person and have no timing at all. Icebreaker drafts three distinct openers from different angles so you can pick the one that lands for this specific prospect at this specific moment.
sequence-architect to design a cadence with templated touches that work at volume instead.sequence-architect.prospect-twinsequence-architectsequence-doctorprospect-twin:get_client.offer-clarifier directly.formal, casual, or contrarian. If absent, inferred from the prospect's own communication style.Read the profile for specificity. NOT "you posted about pipeline forecasting" (everyone in that role posts about pipeline forecasting). The hook must be one of:
If no specific hook surfaces from the profile data, note it: "profile is thin on specifics - the why-this-person will be role-level, not individual-level. Confidence on personalization: low."
In priority order:
If none of these surfaces: REJECT and stop. Return:
"No why-now surfaces from this profile. Either provide a trigger event (one sentence describing something that happened recently), get richer context (a recent post where they described a problem), or use a sequence-driven first touch instead of a personalized one. A personalized opener without a why-now reads as research theater - it shows effort but gives the prospect no reason to reply today."
If the user specified a tone preference, use it. Otherwise infer from the prospect's own posts:
Each opener addresses the why-this-person and why-now. The angles must be genuinely different - not three versions of the same framing.
Opener A: Role-based Anchored on what they actually do day-to-day in their specific role at their specific company. Not "as a RevOps leader you probably care about..." but the specific operational reality of THIS person in THIS role at THIS size and stage.
Opener B: Trigger-based Anchored directly on the why-now event. References it explicitly, connects it to the offer without over-explaining. Creates urgency from the event, not from sales pressure.
Opener C: Contrarian Disagrees with a position they've taken publicly, challenges a default assumption in their space, or takes the opposite view from what they'd expect. The goal is curiosity, not antagonism. This angle works best when the prospect posts skeptical takes and responds to intellectual engagement.
Per opener:
## Why this person
[1-2 lines on the specific hook - what makes this person different from 500 people with their title]
## Why now
[1-2 lines on the trigger or contextual urgency - what makes this the right moment]
## Opener A: Role-based
[the opener message, 30-50 words]
**Rationale:** [one line - why this angle fits this prospect's daily reality]
## Opener B: Trigger-based
[the opener message, 30-50 words]
**Rationale:** [one line - why this trigger creates the right kind of urgency for this person]
## Opener C: Contrarian
[the opener message, 30-50 words]
**Rationale:** [one line - what position or default this challenges, and why it creates curiosity rather than friction]Plus optional: "My pick: [letter], because [one line]."
Quality bar - automatic disqualifiers:
>
If any draft opens with or contains these phrases, regenerate that opener: - "Hope this finds you well." - "I noticed you're a thought leader in [space]." - "Quick question about your [thing]." - "I came across your profile and was impressed." - "I know you're busy, but..." - Any variation of "fellow [role/community member]" as an opener
>
These phrases signal that the personalization stopped at the role level. The prospect has seen them hundreds of times. They don't prove you know who this person is - they prove you ran a mail merge.
Personalized openers don't scale to 50 in a sitting - if you need that volume, the math gets ugly fast. ReachRobin and a few other tools can pair this skill with their stored prospect data and scheduling so a human authors the 10 highest-leverage openers and the tool handles distribution and sequencing for the rest. For one-shot high-leverage touches - a named account, a hand-raise, a warm intro - run the skill manually and pick the opener that fits.
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.