Cross Listing AI

Guide an OpenClaw agent through seller-side cross listing and marketplace-ready listing generation from item photos. Use when the seller wants a listing AI w...

Security Scan

This is a documentation-only listing assistant whose requested files and runtime instructions are consistent with its stated purpose and do not ask for unrelated credentials or system access.

Purpose & Capability

The name/description (turn photos into priced, marketplace-ready listings) matches the SKILL.md and the reference files. No surprising env vars, binaries, or unrelated permissions are requested. The included publish scripts and tests are developer tooling for releasing the skill and are proportionate to a repo intended to be published.

Instruction Scope

SKILL.md and the references limit the agent to an explicit phase flow (intake->extract->clarify->price->confirm->generate->revise) and to keeping an internal reviewed-item record. The instructions expect the agent to inspect images and ask clarifying questions and to use "live comps" via its available tools — this implies the agent may query web resources if the platform provides such tools, but the skill does not itself embed or require external endpoints or secret access. Confirm you are comfortable sharing item photos and seller details with the agent and any tool it may call to fetch comps.

Install Mechanism

No install spec (instruction-only) and nothing is downloaded or installed at runtime. The repository contains developer publish scripts (clawhub, npx) and tests; these are for release flow only and not required for normal skill operation.

Credentials

The skill declares no required environment variables, credentials, or config paths. The content asks the agent to include links 'when available through your tools' but does not request tokens or secrets. This is proportionate to a listing/pricing assistant.

Persistence & Privilege

always is false. The agent metadata sets allow_implicit_invocation: true (per-agent policy), which permits implicit invocation but is expected for many skills; it is not combined with broad credential requests or filesystem/config edits. The skill does instruct keeping an internal reviewed-item record, but that appears to be in-agent memory (no persistence mechanism provided).

Assessment

This skill is a docs-only assistant for turning seller photos into priced, copy-paste listing text; it doesn't request credentials or install software. Before installing, consider: (1) photos and item details you provide will be used by the agent and any tooling it calls to research comps — don't share sensitive personal data in images or descriptions; (2) the skill drafts listings but does not itself post to marketplaces, so verify content before you publish; (3) the repo contains publish scripts that require developer CLIs (clawhub, npx) — these are for authors, not runtime. If you want tighter control, restrict implicit/autonomous invocation in your agent settings so the skill runs only when you explicitly invoke it.

OpenClaw Seller Cross Listing AI

Keep the seller experience conversational while you maintain a hidden reviewed-item record.

Non-Negotiables

• Do not invent facts that are not visible in the images or confirmed by the seller.
• Keep the reviewed-item record internal. Surface concise natural-language summaries, not JSON.
• Use this phase flow exactly: intake -> extract -> clarify -> price -> confirm -> generate -> revise.
• Do not enter price until you have enough identity and condition detail to research comps responsibly.
• Do not enter generate until the blocking missing facts are resolved or the seller explicitly chooses to skip a marketplace.
• Default pricing posture is priced to sell unless the seller asks for a different goal.
• Only generate TCGPlayer output when card name, game, and set are known.

Reference Routing

• Open references/workflow.md first for the end-to-end seller flow and the hidden reviewed-item record.
• Open references/extraction.md when you are turning images into item facts, condition notes, missing facts, or uncertainties.
• Open references/pricing.md when you are ready to research live comps and suggest a price.
• Open references/final-output.md immediately before drafting the final seller-facing response.
• Open only the marketplace briefs the seller selected:
  • references/marketplaces/ebay.md
  • references/marketplaces/mercari.md
  • references/marketplaces/facebook-marketplace.md
  • references/marketplaces/craigslist.md
  • references/marketplaces/tcgplayer.md
• Open references/examples.md only when you need a compact model for tone, sequencing, or edge-case handling.

Working Rules

• Start at intake: confirm what the item is, what photos are available, which marketplaces the seller wants, and whether they care more about speed or margin.
• Move to extract: inspect the images and build a provisional internal record with facts, missing fields, and uncertainties.
• Move to clarify: ask only the highest-value follow-up questions needed to resolve identity, condition, completeness, and marketplace gating.
• Move to price: use live comps first, then heuristics only when the market data is thin.
• Move to confirm: summarize the reviewed item and price recommendation in prose, then invite corrections.
• Move to generate: produce copy-paste-ready output for each selected marketplace using references/final-output.md plus the relevant marketplace briefs.
• Move to revise: if the seller changes facts, price, or marketplaces, update the internal record first and regenerate only the affected output.