agentleFS
jqaisystemsjqaisystems/jqai-ai-skills/skills/outreach-pipeline-designer/SKILL.md

outreach-pipeline-designer

Skill3 starsChanged 6 months agoMITView on GitHub

Design safe, human-reviewed prospecting, lead qualification, enrichment, scoring, outreach draft, and follow-up workflows. Use when the user asks to build an outreach pipeline, prospecting workflow, lead scoring system, sales ops dashboard, market research pipeline, cold email review queue, CRM handoff, or any AI-assisted outreach process that must avoid exposing credentials, private data, prospect lists, raw prompts, scraped exports, or unattended sending.

The file

We split this skill into the rules it gives an agent

11 rules. About 0 are common, 8 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈8

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use this skill to design review-first outreach systems where AI helps search, enrich, score, draft, and prioritize, but a human approves …

    line 10
  2. Very specific:

    Read references/pipeline-blueprint.md and choose the minimum viable pipeline stages.

    line 15
  3. Very specific:

    Read references/scoring-rubric.md before defining score ranges, qualification fields, and next actions.

    line 16
  4. Very specific:

    Read references/review-and-safety.md before recommending sending, exporting, enrichment, scraping, or CRM updates.

    line 17
  5. Distinctive:

    Use fake sample leads and fake companies in examples.

    line 19
  6. Distinctive:

    Use this sequence unless the user gives a stronger domain-specific process:

    line 23
  7. Distinctive:

    Do not design unattended sending.

    line 36
  8. Distinctive:

    Do not provide instructions for evading platform limits, harvesting private contact data, bypassing consent, or scraping behind login walls.

    line 37
  9. Distinctive:

    Do not store or publish real prospect lists, raw exports, phone numbers, emails, credentials, prompt text, or enrichment logs.

    line 38
  10. Distinctive:

    Prefer business-level account qualification before contact-level personalization.

    line 39
  11. Distinctive:

    If the user supplies real leads or private data, summarize the design pattern without copying those records into the output.

    line 56

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Outreach Pipeline Designer

Overview

Use this skill to design review-first outreach systems where AI helps search, enrich, score, draft, and prioritize, but a human approves every message, export, schedule, or system-of-record change.

Workflow

  1. Clarify the audience and offer: who the outreach is for, what problem is being solved, and what channels are allowed.
  2. Read references/pipeline-blueprint.md and choose the minimum viable pipeline stages.
  3. Read references/scoring-rubric.md before defining score ranges, qualification fields, and next actions.
  4. Read references/review-and-safety.md before recommending sending, exporting, enrichment, scraping, or CRM updates.
  5. Produce a pipeline design with source inputs, lead fields, enrichment fields, scoring thresholds, drafting rules, review states, handoff steps, and stop conditions.
  6. Use fake sample leads and fake companies in examples. Never include real prospects, emails, phone numbers, API keys, prompt text, logs, CSV exports, or private CRM data.

Default Pipeline

Use this sequence unless the user gives a stronger domain-specific process:

  1. source: collect candidate accounts or contacts from approved inputs.
  2. normalize: convert inputs into a consistent lead record.
  3. enrich: add public business context and fit signals.
  4. score: apply a transparent qualification rubric.
  5. draft: create a first message or follow-up prompt for review.
  6. review: approve, edit, reject, defer, or request more context.
  7. handoff: export approved records, create tasks, or update a CRM.
  8. follow_up: schedule reminders or next actions after human approval.

Safety Rules

  • Do not design unattended sending. Every email, DM, post, CRM mutation, or export needs human approval.
  • Do not provide instructions for evading platform limits, harvesting private contact data, bypassing consent, or scraping behind login walls.
  • Do not store or publish real prospect lists, raw exports, phone numbers, emails, credentials, prompt text, or enrichment logs.
  • Prefer business-level account qualification before contact-level personalization.
  • Include opt-out, suppression, duplicate detection, and disqualification states in any outbound workflow.
  • Tell the user to check current laws, platform terms, and internal policies before launching outreach.

Output Format

Return:

  1. Pipeline summary.
  2. Stage-by-stage workflow.
  3. Lead record schema using fake fields only.
  4. Scoring rubric with thresholds.
  5. Review queue states and human approvals.
  6. Drafting and personalization rules.
  7. Safety, compliance, and publication boundaries.
  8. Minimal build plan or implementation checklist if the user wants to build it.

If the user supplies real leads or private data, summarize the design pattern without copying those records into the output.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.