Router for the found-vs-deliverable pack. Use when the user says: \"where do I start\", \"is this email sendable\", bare /found-vs-deliverable. Names the next step: score the sample, draft a found row, or run verify. For the scoring itself, see found-vs-deliverable.
Installs into .claude/skills of the current project.
Sponsors
SSkills DirectorySkills Directory
Know which skills are safe — weekly.
SSkills DirectorySkills Directory
Is that skill safe to install?
Are you the author of Found Vs Deliverable Kickoff?
Add the live security badge to your README. It updates with every re-scan.
[](https://www.skillsdirectory.com/skills/cmj-hub-found-vs-deliverable-kickoff)
---
name: found-vs-deliverable-kickoff
description: "Router for the found-vs-deliverable pack. Use when the user says: \"where do I start\", \"is this email sendable\", bare /found-vs-deliverable. Names the next step: score the sample, draft a found row, or run verify. For the scoring itself, see found-vs-deliverable."
argument-hint: "[status | sample]"
allowed-tools: Read Bash(python3 */skills/found-vs-deliverable/scripts/score.py:*)
models: ""
---
# Found vs deliverable — kickoff
Fifteen minutes. One address. Find then verify.
## Start
1. If `$ARGUMENTS` is `sample`, run the good and catch-all examples and show both exits.
2. If `$ARGUMENTS` is `status`, report whether `gtm/verify.json` exists and what the scorer says.
3. Otherwise detect state and route.
## State
| State | Next |
| --- | --- |
| No `gtm/verify.json` | Score the samples first (below), then draft one found row with `sendable: false`. |
| Draft exists, never scored | Run `found-vs-deliverable` with `score`. |
| Score exits 1 | Fix the `- … → …` lines; re-score. |
| Score exits 0, `sendable: true` | Say `Next: /cold-email:cold-email`. |
| Score exits 0, held bucket | Keep the row in its bucket. Do not send. Offer to draft the next found address. |
## Sample loop (the worked example)
```bash
python3 skills/found-vs-deliverable/scripts/score.py --file skills/found-vs-deliverable/examples/verify-good.json
python3 skills/found-vs-deliverable/scripts/score.py --file skills/found-vs-deliverable/examples/verify-catchall.json
```
Good exits 0: verified deliverable, sendable. Catch-all exits 1: never send catch-all as valid. Then write yours to `gtm/verify.json`.
## Welcome
```
Find then verify.
Buckets: deliverable | catch-all | unknown | invalid
Banned: never send catch-all as "valid". Found is not sendable until verify = deliverable.
1. Score the sample (good vs catch-all) ← start here
2. Draft one found row (sendable false)
3. Bring a verify result; set the bucket
4. sendable true only when bucket is deliverable
5. Score gtm/verify.json
Ready to score the sample?
```
## Rules to repeat
- Catch-all, unknown, and invalid stay `sendable: false`.
- `stage: found` stays `sendable: false` until verify returns deliverable.
- This pack does not call a paid verifier. You bring the bucket; the scorer gates send.