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

using-appfollow

Always-loaded discipline for every AppFollow toolkit skill. Read this before calling any AppFollow MCP tool. Enforces the preview→approval→confirm write ritual, the shared MCP error-code vocabulary, and the public-tools-only posture. Other skills reference this file rather than restating the rules.

How do I install this agent skill?

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

Is this agent skill safe to install?

  • Gen Agent Trust Hubpass

    The skill defines operational rules for the AppFollow MCP server, including a feedback system that can transmit session transcripts to the vendor with user consent. It presents an indirect prompt injection surface by ingesting and processing untrusted user review data from external platforms.

  • Socketpass

    No alerts

  • Snykfail

    Risk: CRITICAL · 2 issues

What does this agent skill do?

using-appfollow

You are driving the AppFollow MCP server (the AppFollow tools your session lists) on the user's own OAuth-authenticated session. Every tool call bills the user's own workspace credits. There is no API client and no secret in this toolkit — all data flows through the user's MCP session.

Tool names here are base names (whoami, get_reviews, …). They appear with whatever prefix your host adds (appfollow, plugin_appfollow-ai-toolkit_appfollow, a claude.ai connector name) — match the names, not the prefix.

Session opening (always, every run)

  1. Mint a session_id once — only when submit_feedback is among the tools your session lists. Generate a high-entropy opaque id (e.g. a UUID) at the start of the conversation and pass it as an argument on every AppFollow tool call from here on — whoami, get_credits, and all the rest — not only submit_feedback. One exception: get_server_info takes no arguments at all — never pass session_id to it. The server stamps it on each tool-usage record, so an attached tool-usage log auto-links precisely to this conversation by reference. session_id is not identity — never ask the user for it. Without submit_feedback, mint no session_id and pass none.
  2. Call whoami. On an AUTH_REQUIRED error, stop and tell the user to complete the AppFollow sign-in in their MCP host, then retry.

Elicitation

When the request is ambiguous — which app/collection, date range, store, country, competitor set, or metric — ask one crisp scope-narrowing question and offer a sensible default the user can accept in one word. Never silently guess, never stall. Volunteer the useful next step the user probably wants (an adjacent metric, a competitor cut, a trend window). This keeps friction low and is separate from the write ritual below — those stay mandatory regardless.

Writes: preview → approval → confirm (never skip)

Every write tool is a pair — <name>_preview then <name>:

  1. Call <name>_preview(args) → it returns a single-use, 60-second, args-bound confirmation_token (it validates nothing upstream — a wrong id fails at confirm).
  2. Show the user exactly what will be written; get explicit approval.
  3. Call <name>(args + confirmation_token) with identical args. On CONFIRMATION_REQUIRED (expired/mismatched token) re-preview.

Never auto-confirm. The user must approve each write.

Not every connection can write. If whoami.access.write_tools_available is false, the connection is read-only: do not offer or attempt a write or its _preview (nor any name in access.unavailable_tools); an ACCESS_DENIED on one is the role, not a wrong id. When the user asks for one, say their AppFollow role connects read-only and an account owner or admin can make the change. If access.workspaces is "granted", only those workspaces' apps and collections are reachable — take ids from list_collections / list_apps.

Group runs

For a collection or a multi-app list, resolve the full app set before running. No silent fan-out, and no hard app cap. If INSUFFICIENT_CREDITS surfaces mid-run, stop and emit a partial report rather than failing.

MCP error-code vocabulary (surface verbatim; branch per skill)

AUTH_REQUIRED, ACCESS_DENIED, FEATURE_DISABLED, INSUFFICIENT_CREDITS, RATE_LIMITED, NOT_FOUND, VALIDATION_ERROR, UPSTREAM_ERROR, NETWORK_ERROR, UNKNOWN_TOOL, CONFIRMATION_REQUIRED. ACCESS_DENIED means the session is signed in but the call was refused for this account: most often the collection or app in the call is not the account's — take apps_id from list_collections and ext_id from list_apps — and sometimes a plan feature it lacks (first trap below).

Two traps to respect:

  • FEATURE_DISABLED is an unreliable plan-gate signal. Some plan gates surface as ACCESS_DENIED instead — reviews_semantic without the Semantic Analysis feature, for one. Never branch a degrade path solely on FEATURE_DISABLED.
  • INSUFFICIENT_CREDITS can be a false positive. Once one call costs more than the remaining balance, the gateway blocks credit-spending calls for the whole account, so a cheaper call that follows can be refused with the same code even though the balance covers it. Treat it like any other tool error — the universal rail below applies.

Universal rail: a skill never fails the whole report on a single tool error — skip the affected section, note it, keep going.

Public tools only

Only the AppFollow tools your session lists exist; references/tool-map.md documents them. Never invent or reference an internal tool name, internal host, or internal API path.

Findings hand-off

The report skills emit a findings.json under appfollow.toolkit.findings/v1. The toolkit does not talk to any tracker — their own connected tracker agent can file the findings.

Feedback discipline (optional, free)

When submit_feedback is among the tools your session lists, read references/feedback.md after you finish a meaningful unit of work and before you offer the user a quick rating; otherwise skip feedback entirely.

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/using-appfollow">View using-appfollow on skillZs</a>