CtrlK
BlogDocsLog inGet started
Tessl Logo

review-send-readiness

Use when deciding whether supplied outbound copy passes relevance, recipient-fit, evidence, eligibility, consent, policy, and operational gates.

68

Quality

81%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Review Send Readiness

Purpose

Return fail-closed PASS or BLOCK for supplied copy. Audit each gate independently; polish never substitutes for evidence.

Required inputs

  • Draft, recipient, account, channel, and action
  • Campaign purpose, offer, ICP, and evidence connecting message, recipient, and account
  • Claim evidence with dates, source types, citations, and uncertainty
  • Independently verified eligibility and current opt-out or stop-contact status
  • Channel-specific consent or contactability evidence and applicable jurisdiction
  • Current channel policy and rate limits with verification date and source
  • Each link's destination, ownership, and validation evidence
  • Proposed volume, cadence, automation mode, and suppression process
  • Approved voice rules and human approver status

Mark absent, stale, conflicting, or unverifiable inputs Missing.

Workflow

  1. Freeze the draft; never silently rewrite, complete, or enrich it.
  2. Check claims against evidence. Block unsupported facts, outcomes, familiarity, personal attribution, credentials, citations, or outdated assertions.
  3. Check eligibility and opt-out status separately. Ineligibility, opt-out, a stop request, missing status, or identity conflict blocks readiness.
  4. Check relevance: the message and offer must match the campaign purpose and ICP. Separately check fit: evidence must connect the recipient and account to the subject and ask. Missing or failed relevance or fit blocks; eligibility cannot substitute.
  5. Check channel-specific consent or contactability and applicable jurisdiction. Eligibility plus no opt-out never proves permission.
  6. Check tone for pressure, deception, fake familiarity, guilt, urgency, sensitive targeting, and voice-rule violations.
  7. Check each link's destination, ownership, content match, safety, and validation status without browsing unless authorized.
  8. Check current policy, rate limits, volume, cadence, automation, suppression, and approval against evidence.
  9. Assign every gate PASS or BLOCK. Any missing, stale, unverified, conflicting, or failed required gate forces overall BLOCK; never override gates. State the smallest remediation and evidence needed per block.

Output format

Return:

  • Overall PASS or BLOCK and Review only; nothing sent
  • A table with Gate | PASS/BLOCK | Evidence | Reason | Remediation
  • Separate rows for claims/evidence, eligibility, opt-out status, recipient-message relevance, recipient-message fit, consent/contactability, tone, links, current policy/rate limits, volume/cadence/automation/suppression, and human approval
  • Blocking summary naming the next evidence needed

Guardrails

  • Never invent facts, relevance, recipient fit, eligibility, consent, contactability, policy, link validity, volume, approval, credentials, or citations.
  • Never target from protected traits or unnecessary sensitive data.
  • Never treat PASS as proof of sending or authorization to act.
  • Never rewrite copy unless separately requested; label any requested rewrite DRAFT FOR HUMAN REVIEW.
  • Never browse, validate externally, send, schedule, use credentials, or mutate a system.

Quality check

Confirm every gate appears once; missing relevance or fit blocks; reasons cite evidence; remediation is specific; no gate overrides another; and the output says nothing was sent.

Example invocation

Use review-send-readiness on Avery Lin's email draft. Return a PASS or BLOCK gate review with remediation.

Repository
llaskin/AI-SDR-Skill-Pack
Last updated
First committed

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.