---
name: review-leads
description: >-
  Review the job lead pipeline: read leads/contacts/companies markdown and CSV
  indexes, run report.py, summarize status, rank by fit, suggest next actions.
  Use when asking what's in the pipeline, what to apply to, weekly review, or lead stats.
---

# Review leads

Pipeline dashboard. Start with **`report.py`**, then read markdown for detail.

## Run reports first

```bash
.venv/bin/python scripts/report.py              # full funnel
.venv/bin/python scripts/report.py --today      # compact daily view
.venv/bin/python scripts/report.py --companies  # company-focused
.venv/bin/python scripts/report.py --contacts   # contact-focused
.venv/bin/python scripts/report.py --followups  # overdue follow-ups
.venv/bin/python scripts/report.py --strong     # strong leads not drafted
```

Report buckets:

- Top target companies
- Strong leads not drafted
- Interesting companies needing contacts
- Contacts needing first message
- Drafted but not applied
- Applications needing follow-up
- Messages needing follow-up
- Stale/boring leads to archive

## Read

1. `data/leads.csv`, `data/contacts.csv`, `data/companies.csv` — quick index
2. **`profile/target-thesis.md`** — rank by technical ambition and learning upside
3. Individual `leads/*.md` and `companies/*.md` for top candidates
4. `data/interactions.csv` — recent activity

## Report format for Ryan

Give a scannable summary on top of script output:

### Pipeline snapshot

- Total leads by status (found, researched, applied, …)
- Total by match (strong, moderate, weak, bad)
- By source (linkedin, indeed, …)

### Recommended next actions

3–5 bullets max:

- Pursue first — strong match **and** high technical ambition (name the interesting work)
- Research founder/contact (strong company, no outreach_surface yet)
- Follow up (from `--followups` report)
- Skip or archive — generic SaaS, low novelty, weak learning upside

## Status updates

If Ryan says "archive the Acme one" or "I applied to Stripe":

1. **Applied / messaged / follow-up:** use `log_event.py` (preferred) or edit lead frontmatter `status`
2. **Archive:** edit lead frontmatter `status: archived`
3. Always run `sync_indexes.py` after frontmatter changes

```bash
.venv/bin/python scripts/log_event.py \
  --type application --lead L0001 \
  --summary "Submitted via company portal"
```

## End of review session

```bash
.venv/bin/python scripts/sync_indexes.py
.venv/bin/python scripts/validate.py
.venv/bin/python scripts/report.py --today
```

## Voice

Talk to Ryan like a collaborator: "You've got 3 strong leads at technically ambitious teams — I'd reach out to X first because they're doing [specific hard problem]..."

No corporate dashboard speak. Short paragraphs, real company names.

## Do not

- Re-scrape boards (use `collect-leads` or `company-scout`)
- Tailor resumes here (use `resume-tailor`)
- Invent leads not in repo
- Hand-edit CSV indexes