Skip to content
Back to skills

Builder Data

ASecurity

Query builder reputation data via Talent Protocol API. Get Builder Rank, verify humans, resolve identities (Twitter/Farcaster/GitHub/wallet), search by location/country, get credentials, and enrich with GitHub data.

  • 16 stars
  • 0 votes
  • 0 copies
  • 2 views
  • Added June 7, 2026
developmentbashgitapi

Works with

  • api

Security analysis

A100/100

Pro scans all 4 files and shows the line behind each finding

Scanned June 7, 2026

npx -y skills add modbender/skill-library-mcp --skill builder-data --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Builder Data?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Builder Data
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/modbender-builder-data/badge)](https://www.skillsdirectory.com/skills/modbender-builder-data)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: builder-data
description: Query builder reputation data via Talent Protocol API. Get Builder Rank, verify humans, resolve identities (Twitter/Farcaster/GitHub/wallet), search by location/country, get credentials, and enrich with GitHub data.
---

# Talent Powers

Query professional data from [Talent Protocol](https://talent.app) - a platform that tracks builders

**Use this skill to:**
- Find verified developers by location, skills, or identity (Twitter/GitHub/Farcaster/wallet)
- Check builder reputation (ranks by default, scores only when asked)
- Map Twitter accounts with Wallet addresses
- Verify human identity from a wallet
- Search for builder's credentials (earnings, contributions, hackathons, contracts, etc)
- Check the projects each builder is shipping

## Required Credentials

| Variable | Required | Description | Get it at |
|----------|----------|-------------|-----------|
| `TALENT_API_KEY` | **Yes** | API key for Talent Protocol (read access to profile/identity data) | https://talent.app/~/settings/api |
| `GITHUB_TOKEN` | No | Personal access token for higher GitHub rate limits (60/hr → 5,000/hr) | https://github.com/settings/tokens |

**Base URL:** `https://api.talentprotocol.com`

```bash
curl -H "X-API-KEY: $TALENT_API_KEY" "https://api.talentprotocol.com/..."
```

## Endpoints

| Endpoint | Purpose |
|----------|---------|
| `/search/advanced/profiles` | Search profiles by identity, tags, rank, verification |
| `/profile` | Get profile by ID |
| `/accounts` | Get connected wallets, GitHub, socials |
| `/socials` | Get social profiles + bios |
| `/credentials` | Get data points (earnings, followers, hackathons, etc.) |
| `/human_checkmark` | Check if human-verified (optional, don't use by default) |
| `/scores` | Get ranks (default) or scores (only when explicitly asked) |

## Key Parameters

**Identity lookup:**
```
query[identity]={handle}&query[identity_type]={twitter|github|farcaster|ens|wallet}
```

**Filters (all optional, only use when relevant to the query):**
```
query[tags][]=developer              # filter by tag (developer, designer, etc.)
query[verified_nationality]=true     # only verified nationality
query[human_checkmark]=true          # only human-verified (reduces results significantly)
```

**Sorting:**
```
sort[score][order]=desc&sort[score][scorer]=Builder%20Score
```

**Pagination:** `page=1&per_page=250` (max 250)

## URL Encoding

`[` = `%5B`, `]` = `%5D`, Space = `%20`

## Response Fields

**Default → Ranks (always use unless user asks for scores):**
- `builder_score.rank_position` - Primary rank metric
- `scores[].rank_position` where `slug = "builder_score"` - Latest rank

**Only when user explicitly asks for scores:**
- `builder_score.points` - Score value
- `scores[].points` - Individual score values

- `location` - User-entered location (returned in response)

## Location Filter

**DO NOT USE** `query[standardized_location]=Country` - doesn't work.

**USE `customQuery` with regex:**

```bash
curl -X POST -H "X-API-KEY: $TALENT_API_KEY" -H "Content-Type: application/json" \
  "https://api.talentprotocol.com/search/advanced/profiles" \
  -d '{
    "customQuery": {
      "regexp": {
        "standardized_location": {
          "value": ".*argentina.*",
          "case_insensitive": true
        }
      }
    },
    "sort": { "score": { "order": "desc", "scorer": "Builder Score" } },
    "perPage": 50
  }'
```

See [use-cases.md](references/use-cases.md#by-location-country) for more examples.

## Limitations

- Max 250 per page
- GET only for most endpoints (POST for customQuery)
- Simple `query[standardized_location]` param broken - use `customQuery` regex

## GitHub Enrichment

Get projects/repos via GitHub after resolving username from `/accounts`:

```bash
# 1. Get GitHub username
/accounts?id={profile_id} → { "source": "github", "username": "..." }

# 2. Query GitHub
GET https://api.github.com/users/{username}                           # Profile
GET https://api.github.com/users/{username}/repos?sort=stars&per_page=5   # Top repos
GET https://api.github.com/users/{username}/repos?sort=pushed&per_page=5  # Recent
GET https://api.github.com/users/{username}/events/public             # Commits
GET https://api.github.com/search/issues?q=author:{username}+type:pr+state:open  # Open PRs
```

**GitHub Token (recommended):** Without a token, GitHub limits to 60 requests/hr. With a personal access token, you get 5,000/hr.
- Create one at: https://github.com/settings/tokens → "Generate new token (classic)" → no scopes needed for public data
- Use it: `-H "Authorization: token $GITHUB_TOKEN"`

## References

- [endpoints.md](references/endpoints.md) - Full endpoint docs
- [use-cases.md](references/use-cases.md) - Common patterns
- [github-enrichment.md](references/github-enrichment.md) - GitHub data

Files in this skill

  • SKILL.md4.9 KB
  • references/endpoints.md5.2 KB
  • references/github-enrichment.md2.6 KB
  • references/use-cases.md4.6 KB

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.