Skill instructions
ai-discovery-readiness763 Bytes
View saved version →
---
name: ai-discovery-readiness
description: Use when a user wants to assess supplied AI-discovery signals, prioritize visibility gaps, or summarize dated model-capture evidence.
---
# MentionedOn discovery readiness
Use MentionedOn's public MCP tools only for the evidence the user supplies.
- Use `analyze_discovery_readiness` to score supplied page and entity signals.
- Use `build_visibility_action_plan` to prioritize supplied missing or partial findings.
- Use `summarize_scan_evidence` for dated captures and retain all returned limitations.
- Never claim that a tool fetched a site, ran a paid scan, or established current inclusion in a consumer AI product.
- Keep evidence dates and source labels visible, and separate observations from inferences.
mentionedon-answer-capture2.32 KB
View saved version →
---
name: mentionedon-answer-capture
description: "Prepare a repeatable log and prompt set for actual AI-answer captures supplied by the user, preserving assistant surface, date and missing results."
---
# Record comparable AI-answer evidence
Help the user record what assistants actually said about a business. This workflow can prepare prompts and a log without connecting MentionedOn. It does not query other assistants, start paid scans or create scheduled monitoring by itself. An answer generated in this conversation is not a clean capture from another product.
Use the actual business name, owned domain, buyer questions, service area and relevant assistants. Prepare a small fixed set of mostly unbranded buyer prompts plus useful branded/comparison questions. Use `references/prompt-library.md` when examples would help. Keep prompt wording, assistant surface, account/memory state, location and capture method consistent across sessions; record differences rather than assuming comparability.
Use `references/capture-log-template.md` for the per-answer fields. One row represents one actual answer for one prompt on one surface at a recorded ISO date/time with timezone. Record whether the business is named, its own or business-specific citations, and position only when a real ranking supports it. Keep API, consumer-app and search-AI captures separate. A citation alone is not necessarily a mention.
Use actual pasted answers, screenshots or authorized existing observations. Do not invent, simulate or estimate an answer, citation or rank. Log refusal, sign-in, errors and skipped tests with status error or skipped and a reason; unavailable is not not-mentioned. Preserve deliberate repeated runs rather than rerunning until a favorable answer appears.
If the user requests analysis, `summarize_scan_evidence` at https://mentioned-on.com/mcp accepts the supplied dated observations, currently up to 50 per call. Follow its current input schema and preserve denominator/coverage limitations. Without that connection, label any simple arithmetic as your own tally. A capture describes one observation and does not establish what all customers see or that an edit caused a change.
Agree a prospective recheck cadence when useful and write it in the log. This does not schedule or run later checks. Connection guidance: https://mentioned-on.com/developers/mcp.
Referenced files: 2 — file references are metadata, not package contents.
mentionedon-discovery-evidence1022 Bytes
View saved version →
---
name: mentionedon-discovery-evidence
description: "Use when a user wants to assess supplied AI-discovery signals, prioritize visibility gaps, or summarize dated model-capture evidence."
---
Use the public MentionedOn tools at https://mentioned-on.com/mcp for supplied evidence. If unavailable, give the connection handoff at https://mentioned-on.com/developers/mcp. This workflow does not require access to an owned monitoring account.
# MentionedOn discovery readiness
Use MentionedOn's public MCP tools only for the evidence the user supplies.
- Use `analyze_discovery_readiness` to score supplied page and entity signals.
- Use `build_visibility_action_plan` to prioritize supplied missing or partial findings.
- Use `summarize_scan_evidence` for dated captures and retain all returned limitations.
- Never claim that a tool fetched a site, ran a paid scan, or established current inclusion in a consumer AI product.
- Keep evidence dates and source labels visible, and separate observations from inferences.
mentionedon-visibility-review1.81 KB
View saved version →
---
name: mentionedon-visibility-review
description: "Review a connected MentionedOn business’s saved AI visibility, competitors and comparable changes, or identify the next action from its existing report."
---
# Review saved AI visibility
Use https://mentioned-on.com/mcp through normal OAuth with `brands:read`. If not connected, preserve the task and give https://mentioned-on.com/developers/mcp as the setup handoff; never request credentials in conversation.
Start with `list_monitored_businesses` when the owned business ID is unknown. Select the business the user means; businesses shared by another owner are outside this connection. An empty account gets the returned setup link. Connecting does not start a trial, subscription or scan.
Use `get_visibility_report` for the latest saved capture, competitors, source links and bounded history. Preserve dates, provider failures and coverage. A saved capture does not query consumer AI applications now. Missing or failed captures are unknown, not absence. Compare only the returned complete baseline with matching model, prompt and capture recipe. Do not manufacture a change when comparable baseline fields are null, treat the oldest returned record as a first-ever baseline, or attribute a change to a particular website edit.
Present the main finding, supporting dated evidence and a useful unresolved next action. On a return visit, reread the report and `get_website_fixes` to recognize saved work and installation checks before recommending repeated work. Website drafts and installation verification are separate requested workflows. These reads do not publish a website, run paid scans or notify anyone.
For supplied observations without an owned account, the public evidence tools provide a separate workflow. Access can be revoked at https://mentioned-on.com/oauth/connections.
mentionedon-website-fixes2.06 KB
View saved version →
---
name: mentionedon-website-fixes
description: "Inspect an owned business website, save schema and copy from confirmed facts, and verify that reviewed MentionedOn fixes were installed."
---
# Prepare and verify website fixes
Use the connected MentionedOn service at https://mentioned-on.com/mcp. Normal OAuth needs `brands:read` and, for saved website work, `website:write`. Connection guidance is https://mentioned-on.com/developers/mcp. Tools cover owned businesses only and require an active paid or trial monitoring slot; preserve a returned entitlement handoff.
Find the actual business with `list_monitored_businesses` and read `get_website_fixes` for the current revision, facts, drafts and installation checks. With authorization to inspect the saved public website, call `inspect_business_website` using that exact revision and `authorizationAcknowledged:true`. It saves candidate facts and source receipts while preserving drafts; start inspection before confirming facts. Avoid replacing already reviewed work merely to refresh a page.
Review candidate facts with the user. Call `draft_website_fixes` only for confirmed facts, passing the latest revision and the required confirmation. Leave unknown facts out. Existing user instructions authorizing the save count; do not ask again without a missing decision. Schema, website copy and business-profile text are saved reviewable drafts, not external publication. On a revision conflict, reread and reconcile with the user’s intent before retrying.
The user installs reviewed content in their own website editor. After installation and authorization to check, use `verify_website_fixes` for a page on the same saved website and the exact artifact revision. Preserve dated found, not-found and unknown results. Installation checks do not establish improved AI visibility; a later comparable monitoring capture supplies separate evidence.
These tools do not publish, send messages, change billing or launch paid AI scans. Never promise rankings or current AI inclusion. For connection revocation use https://mentioned-on.com/oauth/connections.
