GitHub AI Tools Intelligence

Done. I scraped GitHub live on August 4, 2026 and saved the results as local CSV/JSON snapshots, along with the scraper script.

It captured 126 tool/repo rows and 34 Marketplace categories, with no fetch errors.

Sources scraped:

• GitHub Marketplace Recommended
• GitHub Marketplace AI Assisted Apps
• GitHub Marketplace AI Assisted Actions
• GitHub topic: ai-tools
• GitHub topic: open-source-ai
• GitHub topic: ai-agents

A few AI tools surfaced today: LovableBot, OpenCode, CodeRabbit, cto.new, Qodo, Claude Code Action Official, AI Code Review Action, browser-use/browser-use, private-gpt, llamafile, firecrawl, langchain, and gemini-cli.

---

Before I create the write ticket, please confirm this placement.

Planned new GoalfyData dataset:
GitHub AI Tools Intelligence

Planned tables:

• github_ai_tools: 126 rows from GitHub Marketplace recommended, AI Assisted apps/actions, and AI-related GitHub Topics.
• github_marketplace_categories: 34 GitHub Marketplace category rows.
• github_scrape_runs: one run record with scrape date, source URLs, row counts, and status.

Why new dataset: your existing datasets are about GoalfyData social/content/outreach, Reddit operations, ecommerce diagnostics, and video workflows. This scrape is a reusable AI-tool discovery/intelligence asset, so a separate dataset is cleaner than mixing it into those.

Reply with confirmed and I’ll create the dataset and import the files.

---

Done. I organized the scrape into a GoalfyData dataset:

GitHub AI Tools Intelligence
Dataset ID: dataset_***
Schema: schema_***

Tables created and loaded:

• github_ai_tools: 126 rows
• github_marketplace_categories: 34 rows
• github_scrape_runs: 1 provenance row

Validation passed: row counts match, key fields have zero nulls, primary keys have no duplicates, and the dataset has registered table metadata, logical relations, governance rules, and a usage guide.

One note: github_ai_tools has 126 source appearances but 119 unique URLs because some repos appear across multiple GitHub Topic surfaces. I preserved those duplicates intentionally so source visibility is analyzable.

---

Dataset Structure
GitHub AI Tools Intelligence
ID: dataset_***
Schema: schema_***
Source: github
Mode: manual import at creation; later enabled managed refresh

Tables

github_ai_tools — 126 rows
Primary key: row_id

column	type	nullable	meaning
row_id	integer	no	stable row number
scrape_date	date	no	scrape date
source	varchar	no	source bucket, e.g. marketplace_ai_apps
surface	varchar	no	normalized surface
kind	varchar	no	app, action, or repository
name	varchar	no	tool/listing/repo name
slug	varchar	no	Marketplace slug or repo name
description	text	yes	GitHub description
stars	varchar	yes	repo stars, blank for Marketplace rows
language	varchar	yes	repo language, blank for Marketplace rows
url	text	no	tool/repo URL
source_url	text	no	page where row was scraped

github_marketplace_categories — 34 rows
Primary key: category_id

column	type	nullable	meaning
category_id	integer	no	stable category row number
scrape_date	date	no	scrape date
slug	varchar	no	Marketplace category slug
name	varchar	no	category display name
url	text	no	category URL

github_scrape_runs — 1 row
Primary key: run_id

column	type	nullable	meaning
run_id	integer	no	scrape run ID
scrape_date	date	no	scrape date
scraped_at_utc	timestamptz	no	scrape completion time
tool_row_count	integer	no	tool row count
category_row_count	integer	no	category row count
source_count	integer	no	number of source pages
status	varchar	no	scrape status
errors_json	text	yes	scrape errors
source_urls_json	text	no	source URL map
local_json_path	text	yes	local JSON snapshot
local_tools_csv_path	text	yes	local tools CSV
local_categories_csv_path	text	yes	local categories CSV
scraper_path	text	yes	scraper script path

Relations

• github_ai_tools.scrape_date → github_scrape_runs.scrape_date
• github_marketplace_categories.scrape_date → github_scrape_runs.scrape_date

Rules

• Preserve source-surface rows: same repo can appear on multiple GitHub Topic pages.
• stars and language apply to repository rows, not Marketplace apps/actions.

---

Done. I created and deployed the dashboard app.

App: github-ai-tools-dashboard
App ID: app_***
URL: private app URL hidden

It is private by default, so opening it redirects through GoalfyData login.

Included:

• Source mix chart
• Repository language chart
• Top repositories ranking
• Marketplace categories
• Searchable explorer
• Detail drawer with GitHub links

Verification passed: local API check, production build, GoalfyData preflight, package upload, and deployment status online.

---

Done. I shared GitHub AI Tools Intelligence with u***@example.com.

Access scope:

• Included: github_ai_tools
• Included: github_marketplace_categories
• Excluded: github_scrape_runs

Share ID: ***
Status: pending
Permission: read-only

The invitation email was sent. If they do not see it, ask them to check spam/junk.

---

Done. I shared the dashboard with v***@example.com.

Share code: ***
Visibility: specified email only
Permission: read-only
Status: pending
Email sent: yes

If they do not see the invitation, ask them to check spam/junk.

---

Done. I shared GitHub AI Tools Intelligence with w***@example.com.

Access scope:

• Included: full dataset

Status: pending
Permission: read-only

The invitation email was sent. If they do not see it, ask them to check spam/junk.

---

I can set the GitHub AI Tools dataset behind the dashboard to refresh automatically every day at 9:00 AM Asia/Shanghai. The dashboard will then show updated data automatically because it reads live from the dataset.

Important: this uses GoalfyData Managed Refresh, and each scheduled run consumes 1 data-update credit. Your current quota had room, but I need your explicit confirmation before enabling a daily schedule.

Please reply confirm daily 9am refresh and I’ll configure it.

---

Done
Enabled GoalfyData Managed Refresh for GitHub AI Tools Intelligence (dataset_***).

Schedule is 0 9 * * *, timezone Asia/Shanghai, so it updates every day at 9:00 AM.

Verified all three tables through live sandbox runs:

• github_ai_tools: 126 rows
• github_marketplace_categories: 34 rows
• github_scrape_runs: 1 row

Confirmed cron_enabled=true on all three tables.

Updated the dataset description/usage guide so it now says the dataset is scheduled, not manual-only.

---

Analysis complete. Here's a summary of what the GitHub AI Tools Intelligence dataset reveals:

Dataset Snapshot

• 126 entries from 6 GitHub surfaces, scraped Aug 5, 2026
• 119 unique tools — 7 appear across multiple surfaces
• 3 tables: github_ai_tools, github_marketplace_categories, github_scrape_runs

Market Composition

Segment	Count
Marketplace AI Actions	20
Marketplace AI Apps	20
Marketplace Recommended Apps	26
Open-source repos: ai-agents topic	20
Open-source repos: ai-tools topic	20
Open-source repos: open-source-ai topic	20

Key Findings

1. Code review dominates AI Marketplace — 15 of 40 AI-specific listings are code review tools, including CodeRabbit, Qodo, CodeAnt, and others. It is the single most crowded AI use case on GitHub.

2. Claude leads in CI agents — Anthropic has 3 dedicated GitHub Actions versus 1 for Google Gemini CLI, suggesting Claude is the go-to model for AI-in-CI workflows.

3. Agent frameworks are the hottest trend — The top 4 repositories by stars are all agent-related: ECC, hermes-agent, firecrawl, and langchain.

4. TypeScript = Python for AI tooling — Both are at 31.7% of open-source AI repos, showing AI has moved beyond Python-only into mainstream developer tooling.

5. Marketplace AI is still young — Only 40 dedicated AI listings across 34 total Marketplace categories, but 5 AI apps already appear in both “AI Assisted” and “Recommended”, signaling crossover traction.

---