What it does
YouTube Analytics Narrator turns YouTube channel performance metrics (views, watch time, CTR, retention, subscriber change) into a readable narrative and data-driven recommendations. Every KPI, period-over-period delta, trend direction, and anomaly signal is computed deterministically in code; an LLM is used only to write the narrative prose and ground the recommendations, never to invent or alter a number. Live YouTube Data/Analytics API access is gated behind an approved OAuth token — until provisioned it runs connector-ready on deterministic mock fixtures. Guardrails: read-only scope, prompt-injection defang of free-text titles and descriptions, no fabrication, no growth guarantees, decision-support framing throughout. Built for YouTube channel owners and content strategists who want trustworthy performance reporting instead of manually reading raw analytics exports.
Example prompts
- Give me a full performance narrative for my channel this month
- What changed in my channel's metrics compared to last period?
- Show me deterministic trends and anomalies from this raw data
- Validate this analytics result for consistency
Tools (8)
Tools the agent exposes — your AI client calls them automatically when it needs them.
- run_full — Run end-to-end: collect channel metrics, compute deterministic KPIs/trends/anomalies, synthesize a performance narrative, return the full result JSON.
- fetch_metrics — Read-only collection of YouTube Analytics metrics for a channel (mock fixtures until a live sandbox key is provisioned). No LLM.
- compute_trends — Deterministic analysis (no LLM) over raw data: KPI delta_pct, trends, top videos, and anomaly signals.
- synthesize_narrative — LLM synthesis step: turn a deterministic analysis into a readable narrative. Narrative only, numbers never change.
- validate_output — Validate an AgentResult: well-formed JSON shape and metric consistency. Returns { valid, issues }.
- list_capabilities — List the agent's static capabilities: available tools, supported languages, and declared credential slots.
- discover_intent — Understand your goal and co-design the exact input via clarifying questions before running.
- plan_inputs — Plan/brainstorm the inputs for a tool: returns questions, schema and a ready-to-edit example.
What you'll need to connect
This agent will ask you for the following. You enter them when you connect — they're encrypted and never shared with the creator.
- Service account JSON · optionalIn Google Analytics, find your property ID under Admin → Property Settings. Then create a service-account JSON key in Google Cloud (IAM & Admin → Service Accounts → Keys → Add key → JSON) and give that service account Viewer access to the GA4 property.Paste the full JSON of your service-account key file (not a file path). FindAgent stores it securely and gives the agent a file path to it.Only sent to: youtube.googleapis.com, oauth2.googleapis.com
- YouTube Channel ID (gated) · optionalDefault YouTube channel id used by the live connector. Leave blank to run on mock fixtures.In Google Analytics, find your property ID under Admin → Property Settings. Then create a service-account JSON key in Google Cloud (IAM & Admin → Service Accounts → Keys → Add key → JSON) and give that service account Viewer access to the GA4 property.Paste the value as a single line.Only sent to: youtube.googleapis.com
- Anthropic API Key · optionalOptional. Leave blank on a sampling-capable client to use the host model with no key.Create a key on the Anthropic Console API keys page (console.anthropic.com → API keys).Paste the value as a single line.Only sent to: api.anthropic.com
How you're protected
FindAgent runs these safety checks on every agent automatically. They're always on and can't be turned off.
- Prompt-injection scanning
Every request is checked for known prompt-injection and jailbreak attempts before the agent runs. This is always on.
- Secret-leak scanning
Every response is scanned for leaked API keys, tokens, and other secrets before it reaches you. This is always on.
