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 agentmods add skills/firecrawl/web-agent/structured-extractionnpx skills add firecrawl/web-agent --skill structured-extractiongit clone --depth 1 https://github.com/firecrawl/web-agentWhat 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 | $0.00032 | $0.00675 |
| Opus 5 | $0.00016 | $0.00338 |
| Sonnet 5 | $0.00006 | $0.00135 |
| Haiku 4.5 | $0.00003 | $0.00068 |
Grade A, and why
structured-extraction 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 2d 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 — 64 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Structured Extraction
Use this skill when extracting data that must match a specific JSON schema.
Strategy by task type
Simple query (single fact or small object)
- Search for relevant results.
- Scrape promising results with a targeted query.
- Build the result object and call formatOutput immediately.
Single target research (one entity, multiple fields)
- Search for relevant URLs.
- Scrape to extract data — stay in the orchestrator unless you have many independent sources (roughly 5+) where parallel workers clearly help.
- Compile findings and call formatOutput.
List of items (array in schema)
- Search/scrape to get the list of items.
- Are all requested details included in the list?
- Yes: Build the result and call formatOutput.
- No: If there are many items (roughly 5+), use spawnAgents so each worker gets the item and fields; otherwise fetch details sequentially in the orchestrator.
- Aggregate all results and call formatOutput.
All items from a website
- Check sitemaps (sitemap.xml, robots.txt) for an easy route to all pages.
- Scrape the entry page. Determine: pagination? Categories? Subcategories?
- For pagination, use interact to click through every page.
- For categories, scrape each category — use spawnAgents only when many independent categories warrant parallel fan-out.
- Aggregate and call formatOutput.
Scraping for structured data
- PREFER scrape with a targeted query over raw page dumps. It keeps context lean.
- When scraping lists, ALWAYS ask about pagination in your query: "How many total results? Is there a next page?"
- For many independent URLs (roughly 5+), spawnAgents can help — each worker gets specific URLs and fields. Fewer URLs: handle in the orchestrator.
- If a scrape returns a 404 or bot-check, do NOT retry. Move on to alternative sources.
Building the output
- Match the schema EXACTLY. Every required field must be present.
- Use null for missing fields — never omit keys.
- Arrays must be arrays even for single items.
- Numbers must be actual numbers, not strings (10.99 not "$10.99").
- Use bashExec with jq to merge data from multiple sources:
jq -s '.[0] * .[1]' /data/part1.json /data/part2.json > /data/merged.json
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.
- 2d ago First seen · 64 lines · 32 tokens per session scan A faf83e9fafe3
structured-extraction is a skill published in the GitHub repository firecrawl/web-agent (1,223 stars, last pushed 4mo ago), licensed MIT. It adds 32 tokens to every session and 675 once invoked, about $0.0002 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
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
brainstorming
You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
babysit-pr
Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…
imagegen
Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…