skillZs
★ LIVE SKILL TAGS ★
>>> LIVE SKILLS INDEX <<<
* OPEN SOURCE *
NO LOGIN, NO TRACKING
※ REAL INSTALL DATA ※
← back to all skills
appfollow/appfollow-ai-toolkit78 installs

search

Find an app on the App Store / Google Play and start tracking it in AppFollow, then hand structured app references to the feedback-analysis and aso-analysis skills. Read-only search + dedupe works on any plan; the add step degrades gracefully when the plan does not allow writes. Trigger on "find app X", "track our app", "add <app> to AppFollow".

How do I install this agent skill?

npx skills add https://github.com/appfollow/appfollow-ai-toolkit --skill search
view source ↗

Is this agent skill safe to install?

  • Gen Agent Trust Hubpass

    The search skill for AppFollow enables app tracking by searching external stores and managing collections. It incorporates safety features such as explicit user approval for account changes and plan-based error handling. The skill possesses an indirect prompt injection surface due to the ingestion of external app metadata, though the risk is mitigated by required human confirmation for write actions.

  • Socketpass

    No alerts

  • Snykwarn

    Risk: MEDIUM · 1 issue

What does this agent skill do?

search — find + track apps

Read using-appfollow first (session opening, write ritual, error codes).

Inputs

  • query: app name (free-text search keywords).
  • store: the short store code — as (App Store) or gp (Google Play) for the search; add_app takes all ten codes (the using-appfollow skill's references/tool-map.md).
  • country (default us), collection (name; created if new), locale.

Flow

  1. Session: whoami (see using-appfollow).
  2. Normalize input. Search takes free-text keywords only — the ASO search endpoints do not accept a store URL or an ext_id. Trim the query to the app name (strip any pasted URL down to the app name/title before searching).
  3. Search (read-only, any plan). Call aso_search_results (optionally aso_search_suggestions, 12 cr). Present numbered candidates. Auto-pick only a single exact-title match; otherwise the user picks.
  4. Resolve collection. list_collections. If the target collection is new, run add_collection_preview → user approval → add_collection.
  5. Dedupe (read-only). list_apps (apps_id = the target collection's collectionId); if the app's extId is already there, record it under skipped_existing and skip the add. list_apps reports store as the long name (itunes = as, googleplay = gp).
  6. Add (write — may be plan-gated). add_app_preview → explicit user approval → add_app with identical args: store (short code), ext_id, apps_id, locale (two-letter country code). Pass is_competitor: true only when the user says the app is a rival (default: one of their own).
  7. Verify. list_apps again; confirm the app's extId now appears; keep the collection's apps_id with it — the report skills address the app by apps_id + ext_id. A wrong add is undone with remove_app_preview → explicit approval → remove_app (same store, ext_id, apps_id). Check get_credits first: with remaining below 200 the removal may be refused, so do not offer it then. Before the approval, tell the user the removal costs 200 credits and show the app's list_apps row (title, store, isCompetitor), or, for an archived app (not listed by list_apps), its ext_id and store. After the removal, re-check with list_apps that the app is gone (for an archived app, the {ok: true} of the confirmed call is the confirmation). Then leave that app out of the emitted added list and count both the add and the remove in credits_spent_estimate (add 0, remove 200 cr).
  8. Emit the JSON block (below).

Read-only degrade path (plan-gated writes)

add_collection requires an Advanced API / Custom plan; neither it nor add_app carries a feature gate — so a plan block does not arrive as FEATURE_DISABLED. It can surface as ACCESS_DENIED, VALIDATION_ERROR (on add_app, a full apps or competitors limit), or UPSTREAM_ERROR. Therefore:

  • Always complete search + dedupe (read-only — works on any plan).
  • Attempt the add_* confirm.
  • On any error code from the add, do not assert a specific code. Report the candidate refs you found, state plainly that the add could not be completed (likely a plan limitation), and list the found refs under not_added so the report skills can still target them (they may already be tracked).

Output (fenced JSON)

{
  "added": [{"store": "as", "ext_id": "389801252", "apps_id": 12345, "collection": "Core apps", "locale": "us", "title": "Example App"}],
  "skipped_existing": [],
  "not_added": [],
  "collection_created": null,
  "credits_spent_estimate": 12
}

Add the canonical catalog link to the repository README so users can inspect current installs and available audits. The publishing guide covers the complete discovery path.

<a href="https://skillzs.dev/skills/appfollow/appfollow-ai-toolkit/search">View search on skillZs</a>