find-product-directories

Drive the ServiceGraph API (https://api.servicegraph.co) to find and rank directories where a founder can submit a software product — SaaS review sites (Capterra, G2, SaaSHub), launch platforms (Product Hunt and its alternatives), app/startup listings, and general software directories — via the product_directory dataset. The catalog has 1,000+ directories, each enriched with Domain Rating (dr), backlink counts, and organic traffic, so you can rank by real SEO value and reach instead of guessing.

This is a "where to launch / where to get backlinks" skill, not a "who to hire" skill. Each row is a directory you submit to, not a firm and not a product. The payoff for the user is twofold: a backlink from a high-authority domain (SEO) and launch-day referral traffic.

Any HTTP client works (curl, fetch, requests). Examples below use curl.

Sibling skills — defer when the niche is specific

If the user's product is specifically one of these, defer to the dedicated skill (its catalog slice and recipes are tuned for it):

• MCP servers ("where do I list my MCP server", "MCP directories") → find-mcp-directories
• AI tools / AI agents / agent skills ("directories for my AI app", "where to list my AI agent", "agent-skills directories") → find-ai-directories

This skill is the umbrella for everything else founders launch — SaaS, web apps, mobile apps, dev tools, startups in general — and is the right pick when the ask spans niches or names none.

MCP server (preferred for authed calls)

If your harness has the ServiceGraph MCP server loaded (recognizable by tool names containing servicegraph), prefer those tools — the harness handles credentials in its own sandbox via OAuth 2.1 + PKCE, so no token enters LLM context. Otherwise use the REST flow below.

API surface (dataset id: product_directory)

Every endpoint requires the bearer (Authorization: Bearer vk_…). There is no anonymous tier.

Cost model. Discovery / validation / search / brief reads are free — including the dr ranking signal. Unlocking a row costs 10 credits and lasts 30 days; it reveals the gated fields: editor_note (the hand-written submission note — how to submit and whether the listing gives a backlink), organic_traffic, and total_visits. Re-fetching an unlocked row within TTL is free.

Auth

Tokens are vk_* API keys minted in the dashboard.

Keep the token out of the LLM context — never read .env* into your context; dispatch every authed call through a shell wrapper.

1. Just try the call through a shell wrapper that sources .env.local:

   ( set -a; [ -f .env.local ] && . ./.env.local; set +a;
     curl -sS -H "Authorization: Bearer $SERVICEGRAPH_API_KEY" \
          'https://api.servicegraph.co/v1/datasets/product_directory/fields' )
   

2. On 401 unauthorized, prompt the user (don't accept the key in chat):

   "Open https://servicegraph.co/profile/api-keys, sign in, click Create key, and copy the vk_… value. Then add SERVICEGRAPH_API_KEY=vk_… to .env.local here (or export it in your shell). Tell me when done. Please don't paste the key into chat."

3. Retry the same call after the user signals ready. A later 401 means the key was rotated/revoked — re-prompt.

Filter DSL

GitHub-search-style.

filter   := orExpr
orExpr   := andExpr ("OR" andExpr)*
andExpr  := notExpr (("AND")? notExpr)*    # whitespace = implicit AND
notExpr  := ("NOT" | "-") notExpr | atom
atom     := "(" filter ")" | predicate
predicate:= IDENT op valueOrList | bareword
op       := ":" | "=" | ">=" | "\x3C=" | ">" | "\x3C"

Four rules that bite:

1. AND binds tighter than OR. a OR b c parses as a OR (b AND c). Use parens.
2. Comma list = OR within one predicate. industry:software_saas,fintech = either.
3. Negation is -x or NOT x. Use saas -crm, not keyword:saas,-crm.
4. Bareword = keyword search. Free-text substring across the directory's name, title, description, listed metadata, and the dir-catalog niche tag (so saas matches dirs whose niche is "SaaS products"). Multiple barewords AND. Wrap multi-word phrases in double quotes ("product launch").

Fields that matter here

This dataset ranks directories by SEO value and reach, so the numeric signals are the point:

Because dr is free and briefs are pre-sorted by it, you can rank a shortlist by authority without spending a single credit — unlock only the ones the user wants submission instructions for.

Identifying rows — apex

Directories are keyed by apex domain (producthunt.com, not www.producthunt.com/posts). Strip user-supplied URLs to the apex before calling :apex endpoints or building unlock batches.

Recipes

A. Launch platforms (Product Hunt alternatives)

User: "Where can I launch my SaaS next week besides Product Hunt?"

GET /v1/datasets/product_directory/search?filter=launch&limit=15
# → cards sorted by dr desc; producthunt.com, uneed.best, peerpush.net, tinylaunch.com, …

Present the top N by dr (free). Tighten to high-authority only:

GET /v1/datasets/product_directory/search?filter=launch+dr>=60&limit=15

B. SaaS review / listing directories for backlinks

User: "High-authority directories to list our B2B SaaS for SEO."

GET /v1/datasets/product_directory/search?filter=saas+dr>=60&limit=20
# → saashub.com, saasworthy.com, financesonline.com, capterra.com (via 'saas' niche), …

Software in general (broader than SaaS):

GET /v1/datasets/product_directory/search?filter=software+dr>=60&limit=20

C. Rank by real reach (organic traffic)

User: "Which directories actually send traffic, not just a backlink?"

organic_traffic is gated but filterable — gate on it to surface high-reach dirs, then unlock to see the numbers:

GET /v1/datasets/product_directory/search?filter=saas+organic_traffic>=50000&limit=15
# → small set of high-traffic dirs. Unlock the user's picks to reveal traffic + how to submit.

D. Vertical / niche product

User: "Directories for our fintech app." / "design-tool listing sites."

Combine a keyword (matched against the niche tag) with the industry refiner, or just lead with the keyword:

GET /v1/datasets/product_directory/search?filter=fintech&limit=15
GET /v1/datasets/product_directory/search?filter=design+dr>=50&limit=15

If a keyword over-narrows, drop it and use industry: (e.g. industry:software_saas, industry:fintech) plus dr>=.

E. Intent translator (let the API draft the filter)

User: "Where do I get our new dev tool in front of developers and pick up backlinks?"

POST /v1/datasets/product_directory/translate-intent
  { "intent": "directories to list a developer tool for backlinks and traffic" }
# → {filter, normalized, count}. Sanity-check the count, then search.

F. Unlock submission instructions for the picks

User has a shortlist and wants to actually submit:

# Present briefs ranked by dr (free). "Unlocking 6 = 60 credits, 30-day TTL —
# this reveals each one's submission note (how to submit, backlink yes/no) and traffic."
POST /v1/datasets/product_directory/unlocks
  { "apexes": ["producthunt.com", "saashub.com", "uneed.best", "..."] }
# → editor_note + organic_traffic + total_visits for each.

The editor_note is the operational payoff — e.g. "Listing doesn't give a backlink but seeds several other directories; submit via CLI, see …". Surface it verbatim so the user knows the effort and the SEO return before submitting.

G. BYO apex list — score directories I already have

User pastes a list of directory domains:

1. GET /v1/datasets/product_directory/:apex per domain — free brief with dr (404 = not in catalog, no charge). Flag misses and rank the hits by dr.
2. User picks N to fully enrich. POST /unlocks with all of them = 10×N credits, single atomic charge; reveals submission notes + traffic.
3. Within 30-day TTL, repeated unlock POSTs are free.

Gotchas

• Rows are directories, not firms or products. If the user wants to hire an agency, route to a find-* agency skill. If they want to find a product listed somewhere ("best CRM for us"), this dataset can't answer that — it lists the directories, not their contents.
• dr is free and briefs are pre-sorted by it — rank and shortlist for zero credits; spend only to reveal submission notes + traffic.
• Gated fields can still be filtered. organic_traffic>=10000 narrows the set even though the value stays hidden until unlock.
• Catalog is global, not US-only, and not B2B-only. Don't refuse non-US or consumer-product asks the way the agency skills do — a directory is a directory regardless of who launches.
• Local / brick-and-mortar business directories are a different dataset (business_directory). If the user wants to list a restaurant, clinic, or local service, this isn't it — point them at the branded servicegraph skill.
• Multi-word phrases must be split or quoted. product launch parses as two AND'd keywords; "product launch" is one phrase.
• Unlock is atomic. POST /unlocks with 6 apexes either charges (up to) 60 credits or leaves balance untouched on 402. Plan the batch.
• Within-TTL re-views are free (was_cached:true).

Errors

JSON envelope: {"error": {"code": "...", "message": "..."}}.

End-to-end example

User: "We're launching our B2B SaaS in two weeks. Find the highest-authority directories to submit to for backlinks, and tell me how to submit to the top five."

# 1. Discover (once per session)
GET /v1/datasets/product_directory/fields?include_values=1

# 2. Validate + scope (free)
GET /v1/datasets/product_directory/check?filter=saas+dr>=60

# 3. Search briefs (free) — already sorted by dr desc
GET /v1/datasets/product_directory/search?filter=saas+dr>=60&limit=20
# → present ranked by dr; note which look like launch platforms vs review sites

# 4. User picks 5. "Unlocking 5 = 50 credits, 30-day TTL — reveals each one's
#    submission note (how to submit + backlink yes/no) and traffic."

# 5. Atomic unlock (charges 50 credits)
POST /v1/datasets/product_directory/unlocks
  { "apexes": ["saashub.com", "saasworthy.com", "producthunt.com", "uneed.best", "financesonline.com"] }

# 6. Surface each editor_note verbatim + traffic so the user can prioritize.