Skip to content
SkillMixed1,225 tokens in use

relationship-intelligence

A workflow that merges lead feeds, counts only verified prior touches and rates opportunities in five independent lanes with evidence.

3,862 stars on the repository MIT Licenseericosiu/ai-marketing-skills · relationship-intelligence
npx skills add ericosiu/ai-marketing-skills -s relationship-intelligence
View source

Install command as written in the repository's README

Skill label

measured from its files

What installing relationship-intelligence costs

26tokens idle
1,225tokens in use
2,738tokens, up to

Idle is what sits in context while the skill is installed. In use is its instructions, loaded when it runs. Up to adds the files its instructions link. Counts are approximate, made offline with cl100k_base.

Model dependence Mixed

Script-ledMixedModel-led

Scripts do part of the work and the model's judgement does the rest.

This describes the skill's structure, how much of it is scripts. It does not say how well it works. Scripts: scripts/touch_summary.py.

Touches

shelldetected

Declared tags come from the skill's frontmatter; detected ones from its scripts, its code blocks and the commands it names in inline code. This is detected, not audited.

How it scores

Quality

How well the skill is written, counted from its own files: its frontmatter, its steps, examples, stated output and constraints, its size and whether every file it ships is one it names.How we score it →
Excellent
  • Frontmatter8/15
  • Structure25/25
  • Examples9/15
  • Output15/15
  • Size15/15
  • Resources9/15

Security

Starts at 100 and loses points for each thing in the skill's files that a reader should look at first.How we score it →
Excellent
  • Runs shell commands−5

    does this on the machine, and the frontmatter does not say so

    SKILL.md, line 92

Repository health

Pre-computed score based on development activity, maintenance, community, maturity, and trend momentum.How we score it →
Good
  • Development activity88
  • Maintenance44
  • Community56
  • Maturity28
  • Momentum40

Relationship Intelligence turns scattered signals about a person into a short answer: who matters, for which opportunity, what already happened, who owns the relationship and what to do next. It keeps people, companies, evidence and interaction events separate, and it counts only verified touches.

It evaluates client, recruiting, partnership, sponsorship and acquisition fit as five separate lanes, each with its own evidence and freshness rules. An offline helper script summarizes deduplicated interaction events.

What You Get

  • A data contract for people, companies, properties, observations and interaction events
  • Rules for linking identities, with merge provenance that can be reversed
  • A touch summary that separates owner-personal, team, automated, inbound and unknown-author events
  • Five fit lanes with S, A, B, C, D and Unknown tiers and freshness requirements
  • A team review view spec with filters for lane, owner, source, date and reply state
  • An evaluation and staged rollout guide: read-only, shadow routing, then a small approved live batch

Common Use Cases

  • Consolidating several lead feeds into one reviewable list of people and prior contact
  • Deciding whether a lead is a client, hire, partner, sponsor or acquisition target
  • Checking what contact already happened before anyone sends outreach
  • Rolling out advisory fields in a CRM with a rollback plan

What Makes It Unique

Most lead enrichment skills score a contact once. This one scores five lanes independently, treats unknown as its own state, keeps coverage gaps visible and refuses to claim no prior contact when a source was not searched. It sets approval and consent boundaries for alerts and CRM writes.

Under The Hood

This is mostly a prose workflow with two reference documents and one small Python helper, touch_summary.py, which reads normalized JSON and summarizes verified interaction events. The helper makes no network calls and writes nothing to a CRM. The skill tells the agent to run the repository’s version check when available, to respect existing telemetry consent and to pass only the skill name, elapsed time, success and version to the telemetry logger. Research steps send the agent to approved professional sources or an authorized enrichment provider, and CRM writes or alerts need the user’s specific approval.

Join founders building with open source

Opinionated takes, migration guides, cost-saving tips, and insights from the open source ecosystem.

Subscribe on SubstackJoin 750+ subscribers