vc-signals

VC Signals

AI-Powered Company Radar for Venture Capital

A skill/plugin for Claude Code, Claude Co-Work, and Codex that turns noisy public internet chatter into a weekly Marathon-style radar: up to 50 qualified companies and OSS projects organized by sector, theme, evidence quality, and next action.

See how it works (visual guide)

---

Current state (June 2026 baseline): VC Signals is now a tiered partner-review system, not just a ranked feed. The blessed current run separates Assign Owner, Partner Review Companies, Review-Worthy Companies, Review-Worthy Market Signals, and Evidence Gap Queue rows. Every weekly run now writes quality-gate.json and renders a Canonical Packet Quality Gate so thin or smoke runs are labeled instead of being dressed up as full-quality radar. A June 2 repeatability check was more sober: three safe validations kept unsafe promotions at 0 and produced useful Partner Review / Market Signal / Evidence Gap rows, but only repeated 1 strict Review-Worthy Company. Product Hunt, X, and manual review now carry source-specific evidence completion plans, but repeatable strict company yield still needs stronger official-domain/founder resolution or structured company metadata for top gaps.

---

The Problem

Every week, thousands of signals about emerging technology trends are scattered across Hacker News, Reddit, X/Twitter, GitHub, blogs, and funding announcements. As a VC, you need to:

• Spot trends before they become consensus
• Separate real signal from hype
• Find investable companies you'd never search for directly
• Track which themes are accelerating vs fading
• Do all of this in under an hour, not a full day of research

VC Signals does this in one command.

---

What You Get

/vc-signals radar all

In a few minutes, you get a weekly all-sector radar like this:

# VC Signals Weekly Radar

## Source-Yield Targets

Assign Owner: 1 / 1-3
Partner Review Companies: 13 / 8-15
Review-Worthy Companies: 8 / 8-15
Review-Worthy Market Signals: 5 / 5-10
Evidence Gap Queue: 12 / 10-15
Unsafe Promotions: 0

## Assign Owner

| Company | Domain | Source | Action | Why |
|---|---|---|---|---|
| Voker | voker.ai | HN launch | assign owner | Strong enough company evidence and launch context to route for partner ownership. |

## Partner Review Companies

| Company | Domain | Grade | Source | Missing Evidence | Recommended Manual Check | Why This May Be Noise |
|---|---|---|---|---|---|---|
| SocialEcho 2.0 | socialecho.net | B | Product Hunt | stage/headcount/funding | Check founders, stage, and company metadata before owner routing. | Real product surface, but still needs company maturity proof. |
| Tabstack Web Research | tabstack.ai | B | Product Hunt | stage/headcount/funding | Check founder/team and company metadata. | Launch quality may not equal venture-scale company quality. |
| R0Y OMNI 1.0 | r0y.xyz | B | Product Hunt | stage/headcount/funding | Check public company metadata and investor context. | Strong product surface, but venture maturity is still unverified. |

## Review-Worthy Market Signals

GitHub stays useful even when it is not a company yet. Rising OSS projects become market signals and generate follow-up searches for companies building in the same theme.

## Evidence Gap Queue

Each blocked row says exactly what is missing: official domain, founder/team, stage/funding/headcount, commercial proof, pricing/docs/careers, or manual LinkedIn/Crunchbase-style check.

Each company or project gets a specific Why On Radar, separate Investment Interest and Evidence Confidence scores, a skeptical Why This May Be Noise, missing-evidence fields, and a next action.

The output is deliberately tiered:

• Assign Owner — strict gate; only rows strong enough for ownership.
• Partner Review Companies — good/decent companies worth inspection, often with one or two clear gaps.
• Review-Worthy Companies — credible product/company rows, not owner-ready yet.
• Review-Worthy Market Signals — OSS/theme movement that tells partners where the market is going, even if there is no company yet.
• Evidence Gap Queue — promising blocked rows with exact manual checks.
• Manual Evidence Queue — a first-class analyst workbench: exact checks, promote-if-found, discard-if-not-found, likely payoff, and unresolved gaps.

Market Sector vs Source Lane: Market Sector is the investment category, such as Cybersecurity or AI Infra. Source Lane is where the evidence came from, such as OSS, Reddit, HN, Grounded Web, or TikTok. An OSS repo can therefore be Market Sector = Cybersecurity and Source Lane = OSS.