Skip to content
Back to search

resale-agent-skill-hub

AI-powered multi-platform C2C resale toolkit — 8 Claude Code skills + MCP server for automating second-hand selling

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/madguyevans-creator--resale-agent-skill-hub/badge.svg)](/listings/madguyevans-creator--resale-agent-skill-hub)

HTML

<a href="/listings/madguyevans-creator--resale-agent-skill-hub"><img src="/api/listings/madguyevans-creator--resale-agent-skill-hub/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

Pick one path. Path A is recommended for most users.

Path A: Claude Desktop (recommended, 5 minutes)

**Step 1: Install**

git clone https://github.com/madguyevans-creator/skillhub.git
cd skillhub && pip install -e .
pip install playwright && playwright install chromium

**Step 2: Get API key (one-time)** Go to [console.anthropic.com](https://console.anthropic.com) → API Keys → create key. Save it once:

echo 'sk-ant-your-key' > ~/.broker/api_key

**Step 3: Configure Claude Desktop** Add to `~/Library/Application Support/Claude/claude_desktop_config.json`:

{
  "mcpServers": {
    "skillhub": {
      "command": "python3",
      "args": ["/path/to/skillhub/broker_core/mcp_server.py"],
      "env": { "BROKER_MOCK_MODE": "true" }
    }
  }
}

**Step 4: Restart Claude Desktop.** Done.

Open Claude Desktop, say "帮我卖掉这双鞋" + upload photo. Claude automatically calls 7 MCP tools in the correct order.

Path B: Claude Code CLI

git clone https://github.com/madguyevans-creator/skillhub.git
cd skillhub && pip install -e .
cp -r skills/* ~/.claude/skills/

Use inside Claude Code: - `/personal-broker` — full pipeline - `/broker-recognize` — analyze product photo - `/broker-price` — price research - `/broker-fuse` — set floor price

Path C: Demo without any accounts (mock mode)

By default `BROKER_MOCK_MODE=true`. Everything works with realistic simulated data — no API key, no platform accounts needed. Photo recognition returns a demo Nike sneaker.

git clone https://github.com/madguyevans-creator/skillhub.git
cd skillhub && pip install -e .
# Done — you're in mock mode. Use Path A or B above.

Going real: turn off mock mode

export BROKER_MOCK_MODE=false
# Then bind platform accounts:
python3 skills/broker-auth/scripts/auth.py --all
# This opens a browser. Log into each platform once. Cookies persist.

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

README.md · README · 182 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. # SkillHub — Smart Resale Agent
  2. A collection of 8 Claude Code skills + an **MCP Server** implementing a **Multi-Platform Smart Resale Agent**: an AI-powered conversational workflow for C2C second-hand resale that reduces the barriers to listing and selling pre-owned goods.
  3. **v2**: Now runs as a standard MCP server (JSON-RPC over stdio) that any MCP-compatible client can call — Claude Desktop, Cursor, VS Code, or custom web frontends.
  4. ## Architecture
  5. ```
  6. skillhub/
  7. ├── broker_core/ # Shared engine (pip install -e .)
  8. │ ├── session_manager.py # Browser login → cookie persistence
  9. │ ├── platform_client.py # Authenticated platform API clients
  10. │ ├── state_manager.py # Listing lifecycle state (~/.broker/listings.json)
  11. │ ├── scheduler.py # launchd (macOS) / cron (Linux) registration
  12. │ ├── audit_logger.py # Append-only Transparency Log (~/.broker/audit.jsonl)
  13. │ └── mcp_server.py # ★ MCP Server — 7 tools via JSON-RPC
  14. ├── guardrails/ # ★ Prompt framework (v2)
  15. │ ├── Omni-Agent-Guardrails.yaml
  16. │ └── README.md
  17. ├── skills/ # 8 Claude Code skills
  18. │ ├── personal-broker/ # Hub: orchestrates the 7-step pipeline
  19. │ ├── broker-recognize/ # Step 1: Photo → product info
  20. │ ├── broker-auth/ # Step 3: Bind platform accounts
  21. │ ├── broker-price/ # Step 2: Authenticated price research
  22. │ ├── broker-card/ # Step 4: Listing cards → auto-publish
  23. │ ├── broker-schedule/ # Step 5: Repricing interval & cron/launchd
  24. │ ├── broker-fuse/ # Step 6: Price Shield (floor price)
  25. │ └── broker-delist/ # Step 7: Auto-detect sale → delist all
  26. ├── tests/ # 41 tests
  27. └── docs/ # Setup guides for MCP clients
  28. ```
  29. ## The 7-Step Pipeline
  30. ```
  31. 📷 broker-recognize → Photo → structured product info
  32. 🔐 broker-auth → Open browser → log in once → session persisted
  33. 🔍 broker-price → Search platforms with session → transparency report
  34. 🛡️ broker-fuse → ⚠️ GLOBAL INTERCEPTOR — set unbreachable floor price BEFORE publishing
  35. 📋 broker-card → Generate cards → validate floor → user confirms → auto-publish
  36. ⏰ broker-schedule → Register launchd/cron for daily repricing checks (floor-gated)
  37. ✅ broker-delist → Scheduler detects sale → auto-delist all platforms (no confirmation)
  38. ```
  39. Each skill can also be invoked standalone (e.g., `/broker-price` for pricing only).
  40. ## Prerequisites
  41. ### ANTHROPIC_API_KEY (required)
  42. `broker-recognize` uses the **Anthropic API** (Claude Vision) to analyze product photos. Set your API key:
  43. ```bash
  44. export ANTHROPIC_API_KEY="sk-ant-..."
  45. ```
  46. Without this key, photo recognition will fail.
  47. ### Python 3.9+
  48. ### Playwright (for platform auth)
  49. ## Quick Start
  50. Pick one path. Path A is recommended for most users.
  51. ### Path A: Claude Desktop (recommended, 5 minutes)
  52. **Step 1: Install**
  53. ```bash
  54. git clone https://github.com/madguyevans-creator/skillhub.git
  55. cd skillhub && pip install -e .
  56. pip install playwright && playwright install chromium
  57. ```
  58. **Step 2: Get API key (one-time)**
  59. Go to [console.anthropic.com](https://console.anthropic.com) → API Keys → create key.
  60. Save it once:
  61. ```bash
  62. echo 'sk-ant-your-key' > ~/.broker/api_key
  63. ```
  64. **Step 3: Configure Claude Desktop**
  65. Add to `~/Library/Application Support/Claude/claude_desktop_config.json`:
  66. ```json
  67. {
  68. "mcpServers": {
  69. "skillhub": {
  70. "command": "python3",
  71. "args": ["/path/to/skillhub/broker_core/mcp_server.py"],
  72. "env": { "BROKER_MOCK_MODE": "true" }
  73. }
  74. }
  75. }
  76. ```
  77. **Step 4: Restart Claude Desktop.** Done.
  78. Open Claude Desktop, say "帮我卖掉这双鞋" + upload photo. Claude automatically calls 7 MCP tools in the correct order.
  79. ### Path B: Claude Code CLI
  80. ```bash
  81. git clone https://github.com/madguyevans-creator/skillhub.git
  82. cd skillhub && pip install -e .
  83. cp -r skills/* ~/.claude/skills/
  84. ```
  85. Use inside Claude Code:
  86. - `/personal-broker` — full pipeline
  87. - `/broker-recognize` — analyze product photo
  88. - `/broker-price` — price research
  89. - `/broker-fuse` — set floor price
  90. ### Path C: Demo without any accounts (mock mode)
  91. By default `BROKER_MOCK_MODE=true`. Everything works with realistic simulated data — no API key, no platform accounts needed. Photo recognition returns a demo Nike sneaker.
  92. ```bash
  93. git clone https://github.com/madguyevans-creator/skillhub.git
  94. cd skillhub && pip install -e .
  95. # Done — you're in mock mode. Use Path A or B above.
  96. ```
  97. ### Going real: turn off mock mode
  98. ```bash
  99. export BROKER_MOCK_MODE=false
  100. # Then bind platform accounts:
  101. python3 skills/broker-auth/scripts/auth.py --all
  102. # This opens a browser. Log into each platform once. Cookies persist.
  103. ```
  104. ## Sandbox / Mock Mode
  105. By default, `BROKER_MOCK_MODE=true` — all platform API calls (search, publish, delist) return realistic simulated data. No real HTTP requests are made. This allows:
  106. - **Demo without platform accounts**: test the full pipeline with mock data
  107. - **Safe development**: no risk of accidentally publishing or modifying real listings
  108. - **Anti-scraping avoidance**: no requests that could trigger rate limits or CAPTCHAs
  109. To use real platform APIs, set `BROKER_MOCK_MODE=false` in your environment. Requires platform accounts bound via `broker-auth`.
  110. ## Transparency Log
  111. Every AI-assisted decision is recorded to `~/.broker/audit.jsonl` — an append-only, immutable JSONL file. You can always audit what the agent did and why.
  112. **What gets logged:**
  113. - Price research results (`log_pricing`)
  114. - Price shield checks — blocked or passed (`log_fuse_check`)
  115. - Listing publishes (`log_publish`)
  116. - Auto-repricing events (`log_repricing`)
  117. - Delist events (`log_delist`)
  118. - Schedule registrations (`log_schedule_registered`)
  119. **How to view:**
  120. ```python
  121. from broker_core import audit_logger
  122. print(audit_logger.format_trail("item_abc123")) # Human-readable timeline
  123. trail = audit_logger.get_audit_trail("item_abc123") # Raw JSON array
  124. ```
  125. When every action is recorded with rationale, there is no "black box" — the user can always audit what happened and why.
  126. ## Key Design Decisions
  127. | Decision | Rationale |
  128. |----------|-----------|
  129. | **Not web scraping** | Uses user's own session cookies. Searching as a logged-in user, not crawling. |
  130. | **Bind once, reuse** | Platform login happens once via browser. Session persisted until expiry (~30 days). |
  131. | **Sold median pricing** | Recommended price = median of actually-sold comparables. Falls back to 5% below active median. |
  132. | **Price shield as global interceptor** | Every price-mutating action (publish, repricing) validates against the floor. Blocked actions are logged to audit trail. |
  133. | **Auto-execute after confirm** | Once user confirms card + floor + schedule, the system auto-publishes and auto-reprices without re-confirmation. |
  134. | **No confirmation on delist** | When scheduler detects sale on any platform, auto-delist all others immediately. |
  135. | **launchd / cron** | System-level scheduling, no daemon process needed. |
  136. | **Mock mode** | `BROKER_MOCK_MODE=true` by default — safe demo without real accounts or anti-scraping risk. |
  137. | **Append-only Transparency Log** | Immutable JSONL log of every decision. The user can always audit what happened. |
  138. ## License
  139. MIT — see [LICENSE](LICENSE)

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.