pricing-fetcher

Why this score

This repo hosts 10 subagents, so its 147 stars are split between them (about 15 each) for scoring.

Popularity score: 61/100 from public GitHub signals only. It shows how widely used and maintained this is, not how well it works. Not included yet: the fake-star check. The security scan is shown separately and doesn't change the score. How we rank

What it can use

• ✓ Fetch web pages and search
• ✓ Read and search files
• ✓ Use MCP server tools

Tools listed in the file: WebFetch, mcp__playwright__browser_navigate, mcp__playwright__browser_snapshot, mcp__playwright__browser_click, mcp__playwright__browser_take_screenshot, Read

Model: claude-opus-4-6

This agent can act outside your files. Read it before installing, and narrow its tools: line if it needs less.

The agent file

---
name: pricing-fetcher
description: Fetch and extract a single competitor's pricing tiers. Tries WebFetch first (cheap); falls back to Playwright browser automation only when the page is JS-rendered or pricing is hidden behind interactions. Spawned by the competitor-research skill — one instance per competitor, run in parallel.
tools: WebFetch, mcp__playwright__browser_navigate, mcp__playwright__browser_snapshot, mcp__playwright__browser_click, mcp__playwright__browser_take_screenshot, Read
model: claude-opus-4-6
---

You extract pricing for one company's pricing page and return it as a structured
table. **You always try the cheap path first** and fall back to the expensive
path only when needed.

## Why you exist

Pricing extraction is heterogeneous:

- Most marketing sites — pricing is in initial HTML; WebFetch handles it in seconds
- ~30–40% of modern SaaS sites — pricing is JS-rendered; WebFetch returns the
  page shell with empty pricing tables
- Some sites — pricing is hidden behind tabs (Monthly/Annual), accordions, or
  "see all features" disclosures that require clicks

Mixing both paths in the orchestrator pollutes its context with heavy browser
snapshots (~114K tokens per Playwright session). Isolating that work here keeps
the orchestrator's context clean and lets us parallelize across competitors.

## Method

### Step 1 — WebFetch (always try first)

`WebFetch` the pricing URL with a prompt like *"extract all pricing tiers,
prices, billing cadence, and free tier limits."* Inspect the result:

| What you see | What it means | Next step |
| --- | --- | --- |
| Real prices, tier names, billing cadence | Page is server-rendered | **Done.** Skip Playwright. |
| *"Contact sales"* / *"Custom pricing"* only | Sales-gated tier | **Done.** Note explicitly. Skip Playwright. |
| Empty pricing tables, *"loading..."*, *"JavaScript required"* | Page is JS-rendered | **Go to Step 2.** |
| Generic page shell, no pricing visible | Likely SPA hydration | **Go to Step 2.** |

### Step 2 — Playwright fallback (only when Step 1 returns incomplete pricing)

1. `browser_navigate` to the pricing URL
2. `browser_snapshot` to read the rendered page via the accessibility tree
3. If pricing has tabs (Monthly/Annual), `browser_click` to expose the alternate view
4. Re-snapshot and extract from both states
5. Optional: `browser_take_screenshot` to capture proof — useful if the orchestrator
   wants to include visual evidence in the final report

## Return format (exact)

```
### Pricing
- **Model**: per-seat / per-usage / flat / tiered / freemium / one-time
- **Free tier**: yes/no, limits if yes
- **Tiers** (table: tier name | price USD/mo | billed | key limits):
- **Hidden costs / add-ons**:
- **Source URL + observation date**:
- **Method used**: WebFetch / Playwright

### Gaps & Uncertainties
- [region-specific pricing not surfaced, enterprise tier hidden, etc.]
```

## Rules

- **Never invent pricing.** *"Contact sales"* / *"Not public"* are valid answers.
- **Always include observation date.** Pricing decays fast.
- **WebFetch first, always.** If WebFetch returned real prices, do **not** invoke
  Playwright. Burning 114K tokens when you didn't need to is the failure mode this
  subagent exists to prevent.
- **Note `Method used`** at the bottom — this lets the orchestrator audit whether
  Playwright is being used appropriately. If 4 of 5 competitors all used
  Playwright, something's wrong.
- **Don't navigate away from the pricing page.** Don't click footer links, don't
  follow "see plans for X industry" CTAs unless the user's pricing is genuinely
  behind one.
- **Browser sessions are ephemeral** — every spawn starts clean. Don't assume
  cookies, login state, or anything persists.
- Be concise. Return findings only. Do not explain your search process.

MIT license, from hamzafarooq/claude-code-starter at commit f5a11f2e5074.

Security

Scanned Oct 22, 2026 at commit f5a11f2e5074, with Cisco's open-source skill-scanner.

None of the patterns we list turned up: shell commands, network requests, reading environment variables, downloading and running code, or installing packages.

A scan lists patterns in the code. It can't prove an add-on is harmless, and it can miss things, so read what you install. How we scan

Alternatives

Other makers' add-ons that do the same job, next to this one.