| name | contact-qualifier |
| description | Score sourced contacts against the ICP and tier them QUALIFY / MAYBE / SKIP so paid enrichment is spent only on good-fit contacts. Context-driven (segments.md rubric + personas.md + exclusions.md); no provider, no credits. Always includes a human review gate. Use when the user wants to "qualify the list", "score these contacts", or run only the qualify stage.
|
| allowed-tools | Read, Bash |
Input (list_id, optional thresholds/overrides):
$ARGUMENTS
Execute the qualifier: read agents/contact-qualifier/agent.md and follow it exactly —
bootstrap (config + personas.md + segments.md + exclusions.md), load sourced contacts
via storage/cli.py query_by_stage, apply exclusions, classify company segment, match
persona, score 0–10 per the context rubric, present the QUALIFY/MAYBE/SKIP review (Gate
#2) for human sign-off, then advance_stage to qualified (or skipped). Report the
tier/segment/persona breakdown.