Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add nospicyplease/amazon-ppc-advanced-skills --skill amazon-search-term-harvest-plannergit clone --depth 1 https://github.com/nospicyplease/amazon-ppc-advanced-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/nospicyplease/amazon-ppc-advanced-skills/amazon-search-term-harvest-planner)<a href="https://agentmods.dev/skills/nospicyplease/amazon-ppc-advanced-skills/amazon-search-term-harvest-planner"><img src="https://agentmods.dev/badge/skills/nospicyplease/amazon-ppc-advanced-skills/amazon-search-term-harvest-planner/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/nospicyplease/amazon-ppc-advanced-skills/amazon-search-term-harvest-planner"><img src="https://agentmods.dev/badge/skills/nospicyplease/amazon-ppc-advanced-skills/amazon-search-term-harvest-planner.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00129 | $0.04975 |
| Opus 5 | $0.00064 | $0.02488 |
| Sonnet 5 | $0.00026 | $0.00995 |
| Haiku 4.5 | $0.00013 | $0.00498 |
Grade A, and why
amazon-search-term-harvest-planner scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 295 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Amazon Search Term Harvest Planner
Purpose
Turn Amazon Ads search term data into a safe, specific harvesting plan. Identify search terms that deserve exact-match isolation, product-target expansion, bid direction, destination routing, or watchlist treatment. Do not treat every converting query as harvest-ready, and do not add source negatives unless routing or waste evidence supports it. Use orders or conversions as the negative-safety signal; treat sales as revenue, not order count.
Optimize for clean traffic control, profitable growth, and learning. Preserve brand defense, own-ASIN defense, launch/rank-defense traffic, and strategically valuable discovery until the data proves a safer route.
This skill works in standalone mode from static exports and in Rocketcart MCP mode when a live Amazon Ads + product-intelligence layer is available. It may prepare approval-ready rows, but it must not create keywords, negatives, product targets, bids, budgets, or campaigns by default.
Connection Modes
Standalone Mode
Use this mode when the user provides pasted tables, CSVs, screenshots, or summaries.
- State that Rocketcart MCP was not used.
- Build the best possible harvest plan from static search-term, targeting, negative, product, and destination data.
- Lower confidence when live entity IDs, current negatives, current bids/states, product context, snapshots, or recent changes are unavailable.
- Do not present rows as executable without live preflight.
Rocketcart MCP Mode
Use this mode when Rocketcart MCP capabilities are available or the user asks for live Rocketcart review, preflight, execution, readback, or monitoring.
Initial Rocketcart review is read-only. Do not use write capabilities until the user explicitly approves exact rows after live preflight.
- If profile is missing, use profile discovery first. If exactly one profile fits, state the assumption; if multiple profiles plausibly match, ask the user to choose.
- Read current campaign, ad group, product-ad, keyword/target, negative, budget, state, bid, and destination context where available.
- Read search-term, targeting, recent-change, snapshot/changelog, and entity-history context where available before trusting stale exports.
- Read product intelligence where available: ASIN/SKU mapping, inventory or availability, Featured Offer / Buy Box, price, reviews/rating, category rank/BSR movement, estimated demand, competitor signals, margin/readiness, and mixed-ASIN risk.
- Use live reads to resolve exact IDs, detect duplicate exacts/targets, detect current negative conflicts, verify destination delivery feasibility, and detect stale rows.
- Produce approval packets for exact candidate rows; execution remains separate.
- After any explicitly approved execution, read back affected entities and define 3/7/14-day monitoring.
- Read Rocketcart search-term harvest mode when you need the Rocketcart-specific read, preflight, approval, execution, and readback sequence.
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 295 lines · 129 tokens per session scan A e0813a95713b
amazon-search-term-harvest-planner is a skill published in the GitHub repository nospicyplease/amazon-ppc-advanced-skills (14 stars, last pushed 3mo ago), licensed MIT. It adds 129 tokens to every session and 4,975 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
zach-product-research
A Sorftime-based product research skill for finding Amazon market opportunities and deciding whether a new product is worth pursuing. Sorftime is a market-research data source for Amazon sellers.
zach-seller-skill-creator
A Chinese-language guide for Amazon sellers who want to turn repeated work processes into reusable skills for an AI agent.
zach-listing-health-checker
An Amazon listing health checker that examines a product page as a shopper would see it, including visibility, price, seller, cart, delivery, category, rank, reviews, and search visibility.
zach-search-term-analyzer
An analyzer for Amazon Brand Analytics Top Search Terms reports, which show popular searches across Amazon and how clicks and conversions are distributed among products.
sealeap-amazon-apparel-lifecycle-ads
A playbook for planning advertising on Amazon US for clothing products, using an ASIN lifecycle. An ASIN is Amazon's product identifier; the lifecycle describes whether a product is new, growing, mature, declining, or seasonal.
zach-search-term-report-analyzer
An Amazon Ads search-term report analyzer for Sponsored Products, Sponsored Brands, and Sponsored Display campaigns. It groups related search terms, measures results over 7, 14, and 30 days, and produces reports in several file formats.