CB booksearch-api
Search Amazon KDP books on the BeyondBSR public API, retrieve BSR (Best Sellers Rank) history for a single book, and explore the Amazon category taxonomy (browse nodes) for any supported marketplace. Each book-search result now ships its full root → leaf category ancestor chain(s) inline, so the skill can aggregate market-opportunity reports by macro/sub category without follow-up calls. Use this skill whenever the user wants to discover, filter, or research self-published or traditionally published books on Amazon by BSR, category, keyword, royalty range, rating, reviews, publication date, binding type, or marketplace (currently only the US and FR Amazon marketplaces are populated with data — more coming soon); when the user wants the BSR timeline of a specific ASIN over the last N days; when the user wants to look up Amazon category codes (browse node IDs), walk the category tree (children, ancestors/breadcrumb), or search categories by name to use as filters in book search; or when the user wants to group/aggregate search results by macro category (e.g. "how many Personal Finance opportunities, broken down by sub-category?"). Typical intents include KDP niche research, low-competition book discovery, sales estimation, royalty/revenue projection, competitor analysis, paperback/hardcover filtering, bulk listing of books matching numeric/textual criteria, single-ASIN BSR history charts, resolving a human-readable category name (e.g. "manga", "self-help") into the Amazon `catId` to pass to `categoryIds` in book search, and producing category-grouped market-opportunity summaries from a single search response. Do not use for price-history timelines, review/rating timelines, or account/user data — only book search, BSR history, and category browsing are exposed.
As a process B 68/100 · Nearly there — weak spots: inputs and preconditions, execution cost, running it twice
What is at stake
Medium-severity findings: the skill is probably honest, but read what alarmed the scanner.
Below is the worst case for this category. The finding here is medium: the guard saw a sign, not a proof.
The instructions or scripts send data out: environment variables, keys, file contents, chat history. You may never notice, because the agent performs the upload as "part of the task".
If the upload is not needed for the task, remove it: catalogs flag such skills and delist them. If it is needed, name the destination explicitly, say what leaves the machine, and give the user a switch.
How to improve
- Shorten the description to 1024 characters.
- The SKILL.md body is over 5,000 tokens: move reference detail into references/ and load it when needed.
- Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
- A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.
Guard findings · 5
✓ No critical or high findings
Medium and low: 5
-
medium Exfiltration
net-credential-useSKILL.md:545Credential used in a network call (verify the destination is the intended service)curl -H "X-API-Key: $BOOKSEARCH_API_KEY" \
-
medium Exfiltration
net-credential-useSKILL.md:549Credential used in a network call (verify the destination is the intended service)curl -H "X-API-Key: $BOOKSEARCH_API_KEY" \
-
medium Exfiltration
net-credential-useSKILL.md:553Credential used in a network call (verify the destination is the intended service)curl -H "X-API-Key: $BOOKSEARCH_API_KEY" \
-
medium Exfiltration
net-credential-useSKILL.md:557Credential used in a network call (verify the destination is the intended service)curl -X POST -H "X-API-Key: $BOOKSEARCH_API_KEY" -H "Content-Type: application/json" \
-
medium Exfiltration
net-credential-useSKILL.md:565Credential used in a network call (verify the destination is the intended service)curl -H "X-API-Key: $BOOKSEARCH_API_KEY" \
Files scanned: 2. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- error
description-longdescription is 1788 chars, limit 1024 - warning
body-longSKILL.md body ≈ 10146 tokens (recommended < 5000); move details to references/ - note
description-budgetdescription takes 1788 of the ~15000-char shared budget for all skills
Process rating: all ten parameters 68/100
- 0Inputs and preconditions. Does not say what the process needs to start
- 30Running it twice. 15 mutating operations with no state check
- 40Execution cost. Instruction body is 10146 tokens: crowds the task out of the window
- 60Tools and files. Uses tools (bash, web, node) that frontmatter does not declare
- 60Result and completion. Output format stated, no completion criterion
- 85Steps. 46 steps, 1 vague phrases
- 100When it triggers. States when to use and when not to
- 100Failures and branches. 9 branches, has a failure section
- 100Consistency. Name and required fields are in place
- 100Progress reporting. Reports progress
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- low 15 top-level sections: this looks like several domains in one skill
- low The response is described with custom markup (3 tags): a typed call is more reliable
Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.
Quality signals
- +3Description length 1788: 120–800 characters recommended
- -5TODO / placeholder text left in the skill
- +1No license
- +2Single-language instructions
- +5Description quotes 2 example trigger phrases
- +4Description says when NOT to use the skill
- +4Structure: 43 headings
- +3Step-by-step instructions: 46 items
- +3Output format is stated explicitly
- +4Has examples (16 code blocks)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 50.