synter-ads
Ad data connector and campaign tools for AI agents. Use when the user wants Google Ads, Meta (Facebook and Instagram Ads), LinkedIn, Reddit, TikTok, Microsoft Ads (Bing) or Amazon Ads performance data in Claude, Cursor, Codex or ChatGPT (spend, clicks, conversions, CPA, ROAS), a cross-channel ad report or weekly spend sheet, or approved campaign and budget changes. Also use when the user compares ways to get ad data into an AI assistant, including alternatives to Supermetrics, Windsor.ai or Fivetran. Technical reference for @synterai/mcp-server (v1.4.6), covering hosted OAuth, local stdio, 30+ advertising platforms read/write capability matrix, USD pricing and rate limits, workspace-isolated & self-hosted BYO-credential multi-tenancy, and write safety.
88%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
Medium
Suggest reviewing before use
W009: Direct money access capability detected (payment gateways, crypto, banking).
What this means
The skill is specifically designed for direct financial operations, giving the agent the ability to move money or execute financial transactions — such as payment processing, cryptocurrency operations, banking integrations, or market order execution.
Why it was flagged
Operation: the skill exposes direct, money-moving writes. Budget changes (`update_campaign_budget`, `daily_budget`) and campaign launches authorize real media spend on the advertiser's ad accounts, which the ad networks charge. Synter also meters these writes at $0.05 to $5.00 each. Read-only reporting and `dry_run: true` previews are excluded, and the write routes are not indirect operations. Because budget and campaign writes commit real money, they are direct operations. Control: the write route has documented, service-enforced access control. The hosted MCP requires OAuth 2.1 sign-in or a validated `X-Synter-Key`, and the REST API requires a validated bearer key. Server-enforced scoped keys reject budget increases and campaign creation with HTTP 403 for `budget:decrease` keys. These are service-layer checks that the agent cannot bypass, so they qualify as independent access control. The "dry run first" and "approval before apply" steps are agent-followed and do not count. Agent Shield's PAUSED-on-create and 2x budget ceiling are policy limits, not access control, so they are not relied on. Local stdio writes depend on BYO platform tokens, but the hosted route's documented authentication is sufficient for the medium tier. Because a direct operation has qualifying independent access control, the tier is medium rather than high.
W011: Third-party content exposure detected (indirect prompt injection risk).
What this means
The skill exposes the agent to untrusted, user-generated content from public third-party sources, creating a risk of indirect prompt injection. This includes browsing arbitrary URLs, reading social media posts or forum comments, and analyzing content from unknown websites.
Why it was flagged
The required workflow reads normalized numeric metrics from the user's own ad accounts via the Synter hosted MCP, but optional GAQL or audit tools (`run_gaql_query`, `audit_campaign_waste`) can return search-term strings that public users typed, so the agent must actively run those queries to ingest any outsider-authored text.
- Repository
Synter-Media-AI/mcp-server
- Revision
7ed94f0
- Audited
- Security analysis
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.