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 BjornMelin/dev-skills --skill firecrawlgit clone --depth 1 https://github.com/BjornMelin/dev-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/bjornmelin/dev-skills/firecrawl)<a href="https://agentmods.dev/skills/bjornmelin/dev-skills/firecrawl"><img src="https://agentmods.dev/badge/skills/bjornmelin/dev-skills/firecrawl/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/bjornmelin/dev-skills/firecrawl"><img src="https://agentmods.dev/badge/skills/bjornmelin/dev-skills/firecrawl.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 | $0.00034 | $0.02484 |
| Opus 5 | $0.00017 | $0.01242 |
| Sonnet 5 | $0.00007 | $0.00497 |
| Haiku 4.5 | $0.00003 | $0.00248 |
Grade A, and why
firecrawl 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 6d 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.
The source is not reproduced here
No licence file
A repository with no LICENSE is all rights reserved by default, so the body is not copied here. The metadata, the measurements and the link are.
What ships with it
23 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.
- agents/openai.yaml 362 B
- evals/task-cases.json 4.4 KB
- evals/trigger-queries.json 2.5 KB
- references/agent.md 1.9 KB
- references/cache-reuse.md 4.2 KB
- references/command-selection.md 3.3 KB
- references/crawl.md 2.4 KB
- references/download.md 1.8 KB
- references/install-auth.md 2.5 KB
- references/interact.md 1.8 KB
- references/maintenance.md 2.4 KB
- references/map.md 1.5 KB
- references/monitor.md 2.6 KB
- references/output-security.md 2.2 KB
- references/parse.md 3.8 KB
- references/recipes.md 7.2 KB
- references/research.md 1.5 KB
- references/schemas.md 4.1 KB
- references/scrape.md 3.5 KB
- references/search.md 2.8 KB
- scripts/firecrawl-cache-index.mjs 41 KB runs code
- scripts/firecrawl-doctor.mjs 5.2 KB runs code
- scripts/firecrawl-help-snapshot.mjs 3.3 KB runs code
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.
- 6d ago Changed · -17 tokens per session e48bbe954fdb
- 9d ago First seen · 255 lines · 51 tokens per session scan A b6b3ba433c8e
firecrawl is a skill published in the GitHub repository BjornMelin/dev-skills (7 stars, last pushed 4d ago), with no licence file. It adds 34 tokens to every session and 2,484 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 2026-09-03.
Other skills, from other repositories
ai-browser-use
A business-diagnosis assistant for marketing, operations, product, and sales questions about how to approach a task.
argent-device-interact
Interact with an iOS simulator, Android emulator, or Chromium (CDP) app using argent MCP tools. Use when tapping UI elements, performing gestures, scrolling/swiping, typing text, pressing hardware buttons, launching apps, opening URLs, taking screenshots, waiting for an element to appear or disappear, or checking…
dev-pain-finder
Scrape real developer pain points for any keyword, technology, or problem space from Reddit, Hacker News, dev.to, and GitHub Discussions simultaneously — then group complaints by theme, score them by frequency and upvote weight, and return a ranked opportunity map showing where developer frustration is high and…
interview-prep
Generate a structured interview preparation guide for any company by scraping real candidate experiences from Glassdoor, Blind, and Reddit in real time using parallel TinyFish agents. Use this skill whenever a user mentions preparing for an interview at a specific company, wants to know what a company's interview…
run-dashboard-e2e-local-changes
Run Playwright E2E (tests/e2e/) against the docker/ all-in-one harness so it reflects LOCAL code changes, not a stale cached image.
test-engineer
Automated E2E testing with Playwright including the auto-fix loop — generate test cases from the UI, run them, fix failures and re-run until passing, then produce a human-readable report. Test until it passes, not just test and report. Drives /toh-test; use whenever tests must be written, run, or made green.