Update to Omneky
Snapshot Oct 27, 2026 · 00:28 UTC · version 2.0.31
Collection source: downloaded plugin package.
Payment or plan references changed
Skill instructions — Now says: this when an agency operator wants numbers or a rollout across several client brands. One job: cross-brand reports and rollouts, one brand at a time. A...
Observed in published text. Live prices and checkout terms have not been verified by this change.
Up to 3 plugins free · no card required.
All fields and original values.
Skill instructions
Oct 24, 2026 · 13:12 UTC
the connected Omneky MCP server. Gate every named tool on this session's `tools/list`. Missing tools → Omneky app. Do not invent a tool. Never ask for internal ids. This skill owns **cross-brand** reports and rollouts. One brand only → `...
Oct 27, 2026 · 00:28 UTC
this when an agency operator wants numbers or a rollout across several client brands. One job: cross-brand reports and rollouts, one brand at a time. A single brand goes to `omneky-analytics-and-catalogue` or the matching creative or lau...
Compare saved observations
Download comparison JSONSKILL.md text changes
− removed · + added
--- before +++ after @@ -15,32 +15,24 @@ # Agency multi-brand work -Use the connected Omneky MCP server. Gate every named tool on this -session's `tools/list`. Missing tools → Omneky app. Do not invent -a tool. Never ask for internal ids. - -This skill owns **cross-brand** reports and rollouts. One brand -only → `omneky-analytics-and-catalogue` or the matching creative / -launch skill. - -Handoff fields: -[handoff-contract](../_shared/references/handoff-contract.md). -Sourced ids: -[preconditions](../_shared/references/preconditions.md). -Acceptance gates: -[creative-quality-gates](../_shared/references/creative-quality-gates.md). -Other intents: -[capability-registry](../_shared/references/capability-registry.md). -Write for: -[marketer-persona](../_shared/references/marketer-persona.md). -Gaps: -[capability-gaps](../_shared/references/capability-gaps.md). +Use this when an agency operator wants numbers or a rollout across +several client brands. One job: cross-brand reports and rollouts, +one brand at a time. A single brand goes to +`omneky-analytics-and-catalogue` or the matching creative or launch skill. + +Gate every tool on this session's `tools/list`; a missing tool → the +Omneky app. Never ask for internal ids. + +Handoff fields: [handoff-contract](../_shared/references/handoff-contract.md). +Sourced ids: [preconditions](../_shared/references/preconditions.md). +Acceptance gates: [creative-quality-gates](../_shared/references/creative-quality-gates.md). +Other intents: [capability-registry](../_shared/references/capability-registry.md). +Write for: [marketer-persona](../_shared/references/marketer-persona.md). ## Goal -Give an agency operator per-client numbers and a safe multi-brand -rollout: one brand at a time, explicit confirmation, no silent -switch. +Per-client numbers and a safe multi-brand rollout. Never switch +brands silently. ## When to use @@ -59,26 +51,23 @@ | Need | Check | If missing | |---|---|---| -| Session | `get_current_user` | Host OAuth reconnect | +| Session | `get_account` | Host OAuth reconnect | | Brand list | `list_brands` | No brands → Omneky app. Do not invent | -| Reporting coverage | Metric tools per `brand_id` | Skip that brand; empty is not zero | -| Metrics | `get_performance_breakdown` / `get_daily_metrics` / `get_performance_movers` | Narrow the window; do not invent ROAS | -| Launch apply | matching `launch_*` / `set_ad_budget` / `set_ad_entity_status` | Per-brand yes first. Default `PAUSED` | +| Reporting coverage | `get_performance_summary` (`include_coverage=true`) per `brand_id` | Skip that brand; empty is not zero | +| Metrics | `get_performance_breakdown` / `get_performance_summary` / `get_performance_movers` | Narrow the window; do not invent ROAS | +| Launch apply | matching `launch_*` tool / `set_ad_budget` / `set_ad_status` | Name the brand first; the host confirms. Status as asked | ## Flow -One short card per brand (spend, ROAS, CTR, mover, next step), -then a one-line roll-up. Never silent-switch brands. Confirm the -brand set once; do not re-ask per metric. - ### weekly-cross-brand-report 1. `list_brands`. Show the name + id list. Confirm the set (or the named subset) before any metric call. 2. For each confirmed `brand_id`: + - `get_performance_summary` (`include_coverage=true`) - `get_performance_breakdown` (winners / spend) - `get_performance_movers` when they asked what changed - - `get_daily_metrics` when they asked for the chart + - `get_performance_summary` when they asked for the chart 3. Thin samples: "too little data to conclude" + the numbers. 4. Deliverable: one short card per client (spend, ROAS, CTR, mover, recommendation) and a one-line roll-up. Confidence = @@ -86,11 +75,9 @@ ### multi-brand-rollout -1. Confirm the brand set. One brand per cycle. -2. Say the brand name and id. Wait for a yes on **that** brand - before generate or launch. A yes for brand A is not a yes - for brand B. -3. Creative / launch follow those skills. Default `PAUSED`. +1. Resolve the brand set from `list_brands`. One brand per cycle. +2. Name the brand and id before generating or launching for it. Never reuse one brand's id for another. +3. Creative / launch follow those skills. Set `campaign_spec.status_on_launch` as asked (no server default). Pass each brand's own `brand_id`. 4. Deliverable: per-brand status table (quoted, generated, launched `PAUSED`, skipped). @@ -102,8 +89,8 @@ to fill the card. `status=timeout` → narrow that brand's window; do not retry the identical call this turn. -`credit_insufficient` → stop that brand; credits are added in the -Omneky app. Continue other brands only after a new yes. +`credit_insufficient` → stop that brand and state the shortfall +(`omneky-billing`). Carry on with the other brands. ## Gaps
Full technical diff · 1 changed fields
changed /skill_md_contents
"---\nname: omneky-agency\ndescription: >-\n Agency multi-brand Omneky work. Use for a weekly cross-brand\n report (per-client summaries) or a multi-brand campaign rollout\n with per-brand confirmation. Never silent-switch brands. Do not\n use for a single-brand ROAS table\n (omneky-analytics-and-catalogue), first-run onboarding\n (omneky-onboarding), competitor analysis\n (omneky-competitor-analysis), A/B tests (omneky-ab-tests), or\n credit balance (omneky-billing). Keywords: weekly cross-brand\n report, per-client summary, multi-brand rollout, agency\n clients, all brands this week.\n---\n\n# Agency multi-brand work\n\nUse the connected Omneky MCP server. Gate every named tool on this\nsession's `tools/list`. Missing tools → Omneky app. Do not invent\na tool. Never ask for internal ids.\n\nThis skill owns **cross-brand** reports and rollouts. One brand\nonly → `omneky-analytics-and-catalogue` or the matching creative /\nlaunch skill.\n\nHandoff fields:\n[handoff-contract](../_shared/references/handoff-contract.md).\nSourced ids:\n[preconditions](../_shared/references/preconditions.md).\nAcceptance gates:\n[creative-quality-gates](../_shared/references/creative-quality-gates.md).\nOther intents:\n[capability-registry](../_shared/references/capability-registry.md).\nWrite for:\n[marketer-persona](../_shared/references/marketer-persona.md).\nGaps:\n[capability-gaps](../_shared/references/capability-gaps.md).\n\n## Goal\n\nGive an agency operator per-client numbers and a safe multi-brand\nrollout: one brand at a time, explicit confirmation, no silent\nswitch.\n\n## When to use\n\n- Weekly cross-brand / per-client report\n- Multi-brand campaign rollout\n- \"All my clients this week\"\n\n## When not to use\n\n- One brand's ROAS / movers → `omneky-analytics-and-catalogue`\n- First-run setup → `omneky-onboarding`\n- Credit ledger → `omneky-billing`\n- Sign-in only → `omneky-getting-started-and-failures`\n\n## Prerequisites\n\n| Need | Check | If missing |\n|---|---|---|\n| Session | `get_current_user` | Host OAuth reconnect |\n| Brand list | `list_brands` | No brands → Omneky app. Do not invent |\n| Reporting coverage | Metric tools per `brand_id` | Skip that brand; empty is not zero |\n| Metrics | `get_performance_breakdown` / `get_daily_metrics` / `get_performance_movers` | Narrow the window; do not invent ROAS |\n| Launch apply | matching `launch_*` / `set_ad_budget` / `set_ad_entity_status` | Per-brand yes first. Default `PAUSED` |\n\n## Flow\n\nOne short card per brand (spend, ROAS, CTR, mover, next step),\nthen a one-line roll-up. Never silent-switch brands. Confirm the\nbrand set once; do not re-ask per metric.\n\n### weekly-cross-brand-report\n\n1. `list_brands`. Show the name + id list. Confirm the set\n (or the named subset) before any metric call.\n2. For each confirmed `brand_id`:\n - `get_performance_breakdown` (winners / spend)\n - `get_performance_movers` when they asked what changed\n - `get_daily_metrics` when they asked for the chart\n3. Thin samples: \"too little data to conclude\" + the numbers.\n4. Deliverable: one short card per client (spend, ROAS, CTR,\n mover, recommendation) and a one-line roll-up. Confidence =\n coverage + sample size per brand.\n\n### multi-brand-rollout\n\n1. Confirm the brand set. One brand per cycle.\n2. Say the brand name and id. Wait for a yes on **that** brand\n before generate or launch. A yes for brand A is not a yes\n for brand B.\n3. Creative / launch follow those skills. Default `PAUSED`.\n Pass each brand's own `brand_id`.\n4. Deliverable: per-brand status table (quoted, generated,\n launched `PAUSED`, skipped).\n\n## Failure and recovery\n\nAuth / 401 → `omneky-getting-started-and-failures`.\nOne brand empty → skip it; do not pull another brand's rows\nto fill the card.\n`status=timeout` → narrow that brand's window; do not retry\nthe identical call this turn.\n`credit_insufficient` → stop that brand; credits are added in the\nOmneky app. Continue other brands only after a new yes.\n\n## Gaps\n\nNo agency inbox, no organic scheduler, no cross-account\nlogin. See\n[capability-gaps](../_shared/references/capability-gaps.md).\n"
"---\nname: omneky-agency\ndescription: >-\n Agency multi-brand Omneky work. Use for a weekly cross-brand\n report (per-client summaries) or a multi-brand campaign rollout\n with per-brand confirmation. Never silent-switch brands. Do not\n use for a single-brand ROAS table\n (omneky-analytics-and-catalogue), first-run onboarding\n (omneky-onboarding), competitor analysis\n (omneky-competitor-analysis), A/B tests (omneky-ab-tests), or\n credit balance (omneky-billing). Keywords: weekly cross-brand\n report, per-client summary, multi-brand rollout, agency\n clients, all brands this week.\n---\n\n# Agency multi-brand work\n\nUse this when an agency operator wants numbers or a rollout across\nseveral client brands. One job: cross-brand reports and rollouts,\none brand at a time. A single brand goes to\n`omneky-analytics-and-catalogue` or the matching creative or launch skill.\n\nGate every tool on this session's `tools/list`; a missing tool → the\nOmneky app. Never ask for internal ids.\n\nHandoff fields: [handoff-contract](../_shared/references/handoff-contract.md).\nSourced ids: [preconditions](../_shared/references/preconditions.md).\nAcceptance gates: [creative-quality-gates](../_shared/references/creative-quality-gates.md).\nOther intents: [capability-registry](../_shared/references/capability-registry.md).\nWrite for: [marketer-persona](../_shared/references/marketer-persona.md).\n\n## Goal\n\nPer-client numbers and a safe multi-brand rollout. Never switch\nbrands silently.\n\n## When to use\n\n- Weekly cross-brand / per-client report\n- Multi-brand campaign rollout\n- \"All my clients this week\"\n\n## When not to use\n\n- One brand's ROAS / movers → `omneky-analytics-and-catalogue`\n- First-run setup → `omneky-onboarding`\n- Credit ledger → `omneky-billing`\n- Sign-in only → `omneky-getting-started-and-failures`\n\n## Prerequisites\n\n| Need | Check | If missing |\n|---|---|---|\n| Session | `get_account` | Host OAuth reconnect |\n| Brand list | `list_brands` | No brands → Omneky app. Do not invent |\n| Reporting coverage | `get_performance_summary` (`include_coverage=true`) per `brand_id` | Skip that brand; empty is not zero |\n| Metrics | `get_performance_breakdown` / `get_performance_summary` / `get_performance_movers` | Narrow the window; do not invent ROAS |\n| Launch apply | matching `launch_*` tool / `set_ad_budget` / `set_ad_status` | Name the brand first; the host confirms. Status as asked |\n\n## Flow\n\n### weekly-cross-brand-report\n\n1. `list_brands`. Show the name + id list. Confirm the set\n (or the named subset) before any metric call.\n2. For each confirmed `brand_id`:\n - `get_performance_summary` (`include_coverage=true`)\n - `get_performance_breakdown` (winners / spend)\n - `get_performance_movers` when they asked what changed\n - `get_performance_summary` when they asked for the chart\n3. Thin samples: \"too little data to conclude\" + the numbers.\n4. Deliverable: one short card per client (spend, ROAS, CTR,\n mover, recommendation) and a one-line roll-up. Confidence =\n coverage + sample size per brand.\n\n### multi-brand-rollout\n\n1. Resolve the brand set from `list_brands`. One brand per cycle.\n2. Name the brand and id before generating or launching for it. Never reuse one brand's id for another.\n3. Creative / launch follow those skills. Set `campaign_spec.status_on_launch` as asked (no server default).\n Pass each brand's own `brand_id`.\n4. Deliverable: per-brand status table (quoted, generated,\n launched `PAUSED`, skipped).\n\n## Failure and recovery\n\nAuth / 401 → `omneky-getting-started-and-failures`.\nOne brand empty → skip it; do not pull another brand's rows\nto fill the card.\n`status=timeout` → narrow that brand's window; do not retry\nthe identical call this turn.\n`credit_insufficient` → stop that brand and state the shortfall\n(`omneky-billing`). Carry on with the other brands.\n\n## Gaps\n\nNo agency inbox, no organic scheduler, no cross-account\nlogin. See\n[capability-gaps](../_shared/references/capability-gaps.md).\n"
Full snapshot data
{
"description": "Agency multi-brand Omneky work. Use for a weekly cross-brand report (per-client summaries) or a multi-brand campaign rollout with per-brand confirmation. Never silent-switch brands. Do not use for a single-brand ROAS table (omneky-analytics-and-catalogue), first-run onboarding (omneky-onboarding), competitor analysis (omneky-competitor-analysis), A/B tests (omneky-ab-tests), or credit balance (omneky-billing). Keywords: weekly cross-brand report, per-client summary, multi-brand rollout, agency clients, all brands this week.",
"included_files": [],
"name": "omneky-agency",
"skill_md_contents": "---\nname: omneky-agency\ndescription: >-\n Agency multi-brand Omneky work. Use for a weekly cross-brand\n report (per-client summaries) or a multi-brand campaign rollout\n with per-brand confirmation. Never silent-switch brands. Do not\n use for a single-brand ROAS table\n (omneky-analytics-and-catalogue), first-run onboarding\n (omneky-onboarding), competitor analysis\n (omneky-competitor-analysis), A/B tests (omneky-ab-tests), or\n credit balance (omneky-billing). Keywords: weekly cross-brand\n report, per-client summary, multi-brand rollout, agency\n clients, all brands this week.\n---\n\n# Agency multi-brand work\n\nUse this when an agency operator wants numbers or a rollout across\nseveral client brands. One job: cross-brand reports and rollouts,\none brand at a time. A single brand goes to\n`omneky-analytics-and-catalogue` or the matching creative or launch skill.\n\nGate every tool on this session's `tools/list`; a missing tool → the\nOmneky app. Never ask for internal ids.\n\nHandoff fields: [handoff-contract](../_shared/references/handoff-contract.md).\nSourced ids: [preconditions](../_shared/references/preconditions.md).\nAcceptance gates: [creative-quality-gates](../_shared/references/creative-quality-gates.md).\nOther intents: [capability-registry](../_shared/references/capability-registry.md).\nWrite for: [marketer-persona](../_shared/references/marketer-persona.md).\n\n## Goal\n\nPer-client numbers and a safe multi-brand rollout. Never switch\nbrands silently.\n\n## When to use\n\n- Weekly cross-brand / per-client report\n- Multi-brand campaign rollout\n- \"All my clients this week\"\n\n## When not to use\n\n- One brand's ROAS / movers → `omneky-analytics-and-catalogue`\n- First-run setup → `omneky-onboarding`\n- Credit ledger → `omneky-billing`\n- Sign-in only → `omneky-getting-started-and-failures`\n\n## Prerequisites\n\n| Need | Check | If missing |\n|---|---|---|\n| Session | `get_account` | Host OAuth reconnect |\n| Brand list | `list_brands` | No brands → Omneky app. Do not invent |\n| Reporting coverage | `get_performance_summary` (`include_coverage=true`) per `brand_id` | Skip that brand; empty is not zero |\n| Metrics | `get_performance_breakdown` / `get_performance_summary` / `get_performance_movers` | Narrow the window; do not invent ROAS |\n| Launch apply | matching `launch_*` tool / `set_ad_budget` / `set_ad_status` | Name the brand first; the host confirms. Status as asked |\n\n## Flow\n\n### weekly-cross-brand-report\n\n1. `list_brands`. Show the name + id list. Confirm the set\n (or the named subset) before any metric call.\n2. For each confirmed `brand_id`:\n - `get_performance_summary` (`include_coverage=true`)\n - `get_performance_breakdown` (winners / spend)\n - `get_performance_movers` when they asked what changed\n - `get_performance_summary` when they asked for the chart\n3. Thin samples: \"too little data to conclude\" + the numbers.\n4. Deliverable: one short card per client (spend, ROAS, CTR,\n mover, recommendation) and a one-line roll-up. Confidence =\n coverage + sample size per brand.\n\n### multi-brand-rollout\n\n1. Resolve the brand set from `list_brands`. One brand per cycle.\n2. Name the brand and id before generating or launching for it. Never reuse one brand's id for another.\n3. Creative / launch follow those skills. Set `campaign_spec.status_on_launch` as asked (no server default).\n Pass each brand's own `brand_id`.\n4. Deliverable: per-brand status table (quoted, generated,\n launched `PAUSED`, skipped).\n\n## Failure and recovery\n\nAuth / 401 → `omneky-getting-started-and-failures`.\nOne brand empty → skip it; do not pull another brand's rows\nto fill the card.\n`status=timeout` → narrow that brand's window; do not retry\nthe identical call this turn.\n`credit_insufficient` → stop that brand and state the shortfall\n(`omneky-billing`). Carry on with the other brands.\n\n## Gaps\n\nNo agency inbox, no organic scheduler, no cross-account\nlogin. See\n[capability-gaps](../_shared/references/capability-gaps.md).\n"
}SHA-256 of public snapshot: d2288481013440a2b84b5fa935ab44e2581c156991717c0412a96d612978b199