{"id":92268,"plugin_id":"plugin_asdk_app_6ae474ea2c1c81918efff2bfa353d0a5","kind":"skill","collection_source":"plugin_package","comparison_source":null,"observed_at":"2026-11-12T00:09:46.346Z","digest":"1825b0f676151647cb2c9c3bd119930032bdb958ed6f56c1246c7e20f26803af","against":null,"payload":{"description":"Select buyer questions worth tracking in Cituna using Search Console and existing evidence, and add, enable, disable, remove or replace tracked prompts when the user asks.","included_files":[],"name":"prompt-tracking","skill_md_contents":"---\r\nname: prompt-tracking\r\ndescription: Select buyer questions worth tracking in Cituna using Search Console and existing evidence, and add, enable, disable, remove or replace tracked prompts when the user asks.\r\n---\r\n\r\n# Select and manage tracked prompts\r\n\r\nRead [connection and evidence rules](../../references/connection-and-evidence.md).\r\n\r\n## Recommend prompts\r\n\r\n1. Resolve the brand with `list_brands`. Its brand `prompts` configuration, when\r\n   returned, is the current set of `{text, enabled}` entries. Read `get_visibility`\r\n   separately for questions measured in the latest saved scan and their engine\r\n   results. That scan can predate configuration changes and is not the current\r\n   prompt editor. Check `whoami` for workspace-wide prompt capacity. If current\r\n   configuration is missing, say it is unavailable rather than assuming it empty.\r\n2. When GSC is connected, use `gsc_query`, `gsc_query_mix` or\r\n   `gsc_striking_distance` to identify observed buyer demand. `gsc_ai_searches`\r\n   can supply conversational-query candidates, but its classifications do not\r\n   prove AI-origin traffic. Respect history coverage and reporting lag.\r\n3. Propose a small set aligned with the brand's actual offering, market and\r\n   language. Include category discovery, relevant use cases, comparisons and\r\n   buyer concerns where the evidence supports them. Avoid leading premises or\r\n   questions designed solely to force a brand mention.\r\n4. Show each proposed prompt, its supporting query/evidence, buyer intent and\r\n   reason to track. Label rewrites or inferred questions as suggestions, never\r\n   literal GSC queries. If GSC is unavailable, use the user's supplied topics and\r\n   clearly disclose that demand is unverified.\r\n5. Deduplicate against the current configured prompts and show the proposed additions and\r\n   capacity implications. A request for recommendations alone does not authorize\r\n   editing the tracked set. If the user asks to add/track specified prompts, apply\r\n   that instruction without an extra approval loop.\r\n\r\n## Apply changes\r\n\r\nUse `set_prompts` with the real brand ID/domain and the exact requested text:\r\n\r\n| User intent | Argument | Effect |\r\n| --- | --- | --- |\r\n| Track additional questions | `add` | Preserves current questions; duplicates ignored |\r\n| Resume questions | `enable` | Re-enables existing questions |\r\n| Stop tracking for now | `disable` | Stops checks while retaining the question |\r\n| Delete specified questions | `remove` | Removes them; destructive |\r\n| Explicitly replace the whole set | `replace` | Deletes questions omitted from the replacement |\r\n\r\nPrefer `disable` for pausing. Use `remove` only for an explicit deletion request.\r\nUse `replace` only when the user explicitly requests a complete replacement and\r\nthe exact final set is established. Never combine `replace` with other actions.\r\nAn empty replacement is refused by this MCP; to pause everything, use `disable`\r\nwith the existing exact questions. Explain which questions will leave tracking\r\nbefore deletion when the user has not already been told. A prompt-config update\r\ndoes not prove that stored historical scans were deleted. Never silently remove\r\nexisting questions to fit a capacity limit. Adding an already-present disabled\r\nprompt does not resume it; use `enable` for that.\r\n\r\nInspect the returned `counts`, `added`, `removed`, `notFound`, and final `prompts`\r\nlist. An `ok: true` response can still include unmatched questions in `notFound`;\r\nreport those as unapplied rather than claiming every requested edit succeeded.\r\nRead `list_brands` again and compare the target brand's saved `prompts` text and\r\nenabled flags. If that read does not expose the configuration, distinguish the\r\ntool's write receipt from an independent verification and point to the app's\r\nprompt settings. Do not repeat a write just to verify it.\r\n\r\nUse `get_visibility` to check measurements only. New or resumed prompts may not\r\nhave answers until the next scan; changed configuration does not rewrite the\r\nprevious scan. Do not report citations before measurements exist or trigger a\r\npaid scan simply to fill the new grid.\r\n\r\n## Relevant tools\r\n\r\n`whoami`, `list_brands`, `get_visibility`, `get_engine_answers`, `set_prompts`,\r\n`gsc_status`, `gsc_query`, `gsc_query_mix`, `gsc_striking_distance`,\r\n`gsc_ai_searches`.\r\n"},"changes":[],"summary":"First saved snapshot. No earlier version is available for comparison.","summary_kind":"deterministic","summary_metadata":{}}