qualify-leads
Skill298 starsChanged 6 months agoMITView on GitHub
Run the GTM-OS 7-gate qualification pipeline over a batch of leads from a CSV, JSON, Notion DB, profile-visitors export, post-engagers export, or an existing SQLite result set. Writes per-lead pass/fail data back into SQLite and returns a result-set id you can hand to the campaign launcher. Use when the user says 'qualify these leads', 'score this lead list', 'run the qualification pipeline', 'check if these leads are a fit', or 'qualify the engagers'. Side-effecting — writes to the local SQLite db.
- Reads credentials
In this repository37 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- answer-linkedin-commentsSkill
- build-routineSkill
- campaign-dashboardSkill
- campaign-strategySkill
- claap-weekly-recapSkill
- cross-platform-schedulerSkill
- debuggerSkill
- enrich-with-signalsSkill
- find-lookalikesSkill
- fullenrich-content-engagersSkill
- fullenrich-event-attendeesSkill
- fullenrich-network-activationSkill
- fullenrich-plg-reverse-lookupSkill
- icp-importSkill
- import-leadsSkill
- integrate-with-yalcSkill
- launch-linkedin-campaignSkill
- lemlist-campaign-from-icpSkill
- list-adaptersSkill
- lost-deal-revival-agentSkill
- notion-reviewSkill
- personalize-messageSkill
- predictleads-dashboardSkill
- predictleads-lookalikesSkill
- predictleads-signalsSkill
- prospect-discovery-pipelineSkill
- provider-builderSkill
- research-prospectSkill
- run-competitive-intelSkill
- run-doctorSkill
- scrape-post-engagersSkill
- send-cold-emailSkill
- setupSkill
- show-routineSkill
- track-campaignsSkill
- yalc-orientationSkill
The file
We split this skill into the rules it gives an agent
6 rules. About 0 are common, 4 are distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 13
Use this skill when the user says:
- Distinctive: line 46
Onboarding interruption guard: Run:
- Very specific: line 50
If BLOCKED, stop. Tell the user: "Setup is mid-flight.
- Distinctive: line 68
Any toggles?: Ask only if volunteered:
- Distinctive: line 110
Don't run anything unless the user says yes.
- Very specific: line 114
When leads:qualify exits non-zero, paste its stderr unchanged.
≈ “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.
Files
Qualify Leads
I'll wrap leads:qualify. I'll ask for the input source and path, validate locally, run the CLI, parse the per-gate counts, and surface the result-set id so you can hand it to launch-linkedin-campaign or personalize-message.
When This Skill Applies
Use this skill when the user says:
- "qualify these leads"
- "score this lead list"
- "run the qualification pipeline"
- "check if these leads are a fit"
- "qualify the engagers"
NOT this skill (use personalize-message instead):
- "personalize a message for this lead" — that wraps the
personalizeCLI and writes a single LinkedIn DM, it doesn't run the gate pipeline. - "draft a DM for [lead]" — same.
NOT this skill (use scrape-post-engagers instead):
- "scrape engagers off this LinkedIn post" — that wraps
leads:scrape-post. Run that first to produce an engagers JSON, then come back here to qualify it.
What This Skill Does
- Asks where the leads live (CSV path, JSON path, Notion DB id, visitors export, engagers export, or an existing result-set id).
- Validates the input locally — file exists, Notion id has the right shape, result-set id format looks plausible.
- Shells out to
npx tsx src/cli/index.ts leads:qualify <args>from~/Desktop/gtm-os/. - Parses the CLI's stdout — the pipeline emits per-gate counters and the resultSetId.
- Renders a clean per-gate summary, the result-set id, and the top hot leads.
- Offers two follow-up moves: launch a campaign or personalize messages.
What This Skill Does NOT
- Send any outbound messages. That's
launch-linkedin-campaign(LinkedIn) orsend-cold-email(email). - Author personalized message bodies. That's
personalize-message. - Re-import data that's already a result set. If the user already has a resultSetId, pass
--result-set <id>and skip--source/--input. - Modify any
.envfile. If a key is missing, the CLI raises and we surface its stderr verbatim.
Pre-flight (do this before step 1)
- Onboarding interruption guard. Run:
Iftest -f ~/.gtm-os/.in-flight-setup && echo "BLOCKED" || echo "OK"BLOCKED, stop. Tell the user: "Setup is mid-flight. Finishyalc-gtm startfirst, then re-invoke me." Exit cleanly. - Confirm cwd. All shell-outs assume
~/Desktop/gtm-os/.
Workflow
Step 0 — Ask for input
One question at a time:
- Where do the leads live? Five options:
- CSV path (
--source csv --input <path>) - JSON path (
--source json --input <path>) - Notion DB id (
--source notion --input <db-id>) - Profile-visitors export (
--source visitors --input <path>) - Post-engagers export (
--source engagers --input <path>) - Or "use the most recent result set" / a specific id (
--result-set <id>, no--source/--input)
- CSV path (
- Any toggles? Ask only if volunteered:
--dry-run— preview without writing--no-dedup— skip dedup gate--slack-confirm— Slack for ambiguous dedup matches--enrich-signals(+--signals-types <jobs,funding,tech,news>) — pull PredictLeads after qualify
Step 1 — Validate the input locally
- For
csv/json/visitors/engagers:test -f <path>returns 0. - For
notion: 32-char hex (with or without dashes). Extract id from URL if needed. - For
--result-set <id>: format looks plausible.
Step 2 — Shell out
cd ~/Desktop/gtm-os && set -a && source .env.local && set +a && \
npx tsx src/cli/index.ts leads:qualify --source <type> --input <path>…or for an existing result set:
cd ~/Desktop/gtm-os && set -a && source .env.local && set +a && \
npx tsx src/cli/index.ts leads:qualify --result-set <id>Per the merged 0.13.0 benchmark in docs/skills-architecture.md, single-command side-effecting skills shell out unconditionally. The CLI is the source of truth for env loading, tenant resolution, and the withDiagnostics() wrapper.
Step 3 — Parse the output
leads:qualify writes [qualify] log lines to stdout, one per gate, plus the result-set id and survivor count. On non-zero exit, surface stderr verbatim — no summarising.
Step 4 — Render
Show the per-gate funnel, result-set id, and top 5 hot leads. See references/example-output.md.
Step 5 — Offer follow-ups
"Want me to also: (a) launch a LinkedIn campaign for the qualified leads via
launch-linkedin-campaign, or (b) personalize messages for the top N viapersonalize-message?"
Don't run anything unless the user says yes.
Failure surfacing — verbatim, never summarised
When leads:qualify exits non-zero, paste its stderr unchanged. The CLI's messages are tested and stable.
Notes
- Pipeline writes to
~/.gtm-os/gtm-os.db. Side-effecting → Pattern A (shell-out) per the architecture doc. - 7 gates: dedup → prequal → exclusion → company-fit → role-fit → ICP scoring → optional signal enrichment. See
src/lib/qualification/pipeline.ts. --enrich-signals≈ 1 PredictLeads credit per surviving domain (cached 7 days).
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
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.

