meta-ads-audit — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited meta-ads-audit (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
Comprehensive Meta Ads (Facebook + Instagram) account audit using live data via Cogny's MCP server.
Requires: Cogny Agent subscription ($9/mo) — Sign up
Meta is primary. The same audit (structure, fatigue, audiences, pacing) runs on any other paid social account the workspace has connected.
| Job | Meta (primary) | TikTok | Snapchat | X | |
|---|---|---|---|---|---|
| Accounts | tool_list_ad_accounts | tool_list_advertisers | tool_list_organizations → tool_list_ad_accounts(organization_id) | tool_list_ad_accounts | tool_list_ad_accounts |
| Structure | tool_get_campaigns, tool_get_ad_sets, tool_get_ads | tool_get_campaigns, tool_get_ad_groups, tool_get_ads (advertiser_id) | tool_get_campaigns, tool_get_ad_squads, tool_get_ads | tool_get_campaigns, tool_get_ad_groups, tool_get_ads | tool_get_campaigns, tool_get_line_items, tool_get_promoted_tweets (account_id) |
| Performance | tool_get_insights(level, date_preset) | tool_get_insights(start_date, end_date, level: "AUCTION_CAMPAIGN" / "AUCTION_ADGROUP" / "AUCTION_AD") | tool_get_stats(entity_id, start_time, end_time, level) | tool_get_reporting(start_date, end_date, level) | tool_get_analytics(entity, entity_ids, start_time, end_time) |
| Audiences / tracking | tool_get_custom_audiences, tool_get_pixels | tool_get_audiences, tool_get_pixels | tool_list_segments, tool_list_pixels | tool_get_custom_audiences, tool_get_conversion_events | tool_get_tailored_audiences, tool_get_web_event_tags |
| B2B lead quality | Fibbler get_meta_ads_campaign_analytics | — | — | — | — |
Tool names omit the mcp__cogny__mcp__<Prefix>__ part. What differs:
frequency metric) and Reddit (FREQUENCY field). On Snapchat and X, request reach/uniques fields if available, otherwise judge fatigue by CTR decay over time instead.entity_ids.Call mcp__cogny__list_integrations. If Meta Ads is connected, run the steps below. If only TikTok, Snapchat, Reddit or X is connected, run the same steps with that platform's tools from the table and title the report with the platform name. If none is connected:
This skill requires a paid social integration (Meta, TikTok, Snapchat, Reddit or X Ads) via Cogny's MCP server.
Sign up at https://cogny.com and connect your ad account./meta-ads-audit — full audit /meta-ads-audit audiences — audience analysis only /meta-ads-audit creatives — creative performance only /meta-ads-audit budget — budget pacing only
mcp__cogny__mcp__Meta_Ads__tool_list_ad_accountsmcp__cogny__mcp__Meta_Ads__tool_get_campaigns(ad_account_id, status_filter: ["ACTIVE"])
mcp__cogny__mcp__Meta_Ads__tool_get_insights(ad_account_id, level: "campaign", date_preset: "last_30d")mcp__cogny__mcp__Meta_Ads__tool_get_ad_sets(ad_account_id)
mcp__cogny__mcp__Meta_Ads__tool_get_insights(ad_account_id, level: "adset", date_preset: "last_30d")Check: audience overlap, frequency saturation, demographic breakdown
mcp__cogny__mcp__Meta_Ads__tool_get_ads(ad_account_id)
mcp__cogny__mcp__Meta_Ads__tool_get_insights(ad_account_id, level: "ad", date_preset: "last_30d")Flag: fatigue (frequency > 3), low CTR, single-creative ad sets
mcp__cogny__mcp__Meta_Ads__tool_get_custom_audiences(ad_account_id)
mcp__cogny__mcp__Meta_Ads__tool_get_pixels(ad_account_id)Check: audience sizes, stale audiences, pixel health
For B2B lead-gen accounts, Meta's leads say nothing about whether they became pipeline.
mcp__cogny__mcp__app-fibbler-co__get_meta_ads_campaign_analytics(time_period: "90d")
mcp__cogny__mcp__app-fibbler-co__get_deal_attribution_summary(source: "META_ADS", time_period: "90d")Flag campaigns with cheap CPL but no influenced pipeline, and a high untaggedVisits count (missing UTMs caps attribution). Fibbler's "engagements" are not conversions, and its pipeline is influence, not credit; show it next to Meta's numbers, never added to them.
Compare daily budget vs actual spend for last 7 days. Flag:
[Meta / TikTok / Snapchat / Reddit / X] Ads Audit: [Account Name] ([currency])
Health Score: X/100
Account Overview (30d):
Spend: [X] | Reach: [X] | Leads: [X] | CPL: [X]
Influenced pipeline (Fibbler, if connected): [X]
Active Campaigns: [N]
Top Performer: [Campaign] — [CPL/ROAS]
Worst Performer: [Campaign] — [why]
Issues Found: [N]
- [Critical] ...
- [High] ...
Top 3 Actions:
1. ...
2. ...
3. ...Use create_ticket when the workspace has it (Cloud) and create_finding otherwise (Solo). The fields differ; see section 2 of ../../skills/marketing-engineer/references/mcp-routing.md.
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.