Ground an AI Agent in Brand Memory Before Marketing Analysis
Let an external AI agent read Soku brand memory before analyzing campaigns, then verify every performance claim against current ads or analytics data instead of treating memory as live truth.
Best for
Teams that want consistent brand-aware analysis across external agents without pasting positioning, KPI targets, and reporting preferences into every prompt.
Access mode
Read-only analysis
No connected account changes.
The prompt
Read https://soku.ai/cli/skill.md, connect to my Soku workspace, and inspect brand memory for positioning, KPI targets, audiences, and reporting preferences before the analysis. Treat memory as background context, not live measurement. Verify every current performance claim with ads or analytics commands, cite the source window, and flag memory that conflicts with fresh data. Keep the workflow read-only. Safety boundary: do not change connected accounts or publish content. Use read commands only, and return proposed actions for a human to review separately.
Paste this into Claude Code, Codex, Cursor, or another shell-based agent. It will pause when browser approval or a human review is required.
What you will get
The agent uses durable brand context to frame the analysis while grounding every time-sensitive performance conclusion in fresh ads or analytics data and flagging stale memory.
- A brand-aware analysis with memory references and current data citations
- A list of stale or conflicting memory items that need human correction
Before you start
- Relevant positioning, KPI, audience, or reporting references stored in Soku memory
- At least one live data source available to verify time-sensitive claims
Works with
Run it with Soku CLI
- 1
Discover relevant memory
Search for only the references needed by the task instead of loading unrelated brand history.
soku memory search 'positioning KPI audience reporting preferences' - 2
Read the selected references
Retrieve the named memory records and retain their type and age as provenance.
soku memory list - 3
Verify live claims
Use current ads or analytics commands for every metric and compare them with targets from memory.
soku ads --help && soku ga4 --help - 4
Report context conflicts
Call out stale or conflicting memory explicitly rather than silently choosing one source.
Safety boundary
Read onlyMemory is contextual and may be stale. The agent must verify time-sensitive facts against live data, expose conflicts, avoid treating preferences as measured outcomes, and make no account changes.
Related Soku skills: soku-ads-report
More Agent Workflows use cases
Audit Google Ads With Claude Code and Soku CLI
Give Claude Code read-only access to Google Ads through Soku CLI, inspect tracking, search terms, creative coverage, budgets, and Quality Score, then save an evidence-backed account audit.
Review Meta Ads Performance With Codex and Soku CLI
Connect Codex to Meta Ads reporting through Soku CLI, compare campaign and creative performance, identify fatigue and CPA changes, and return a read-only optimization brief with evidence.
Build a Weekly Cross-Channel Ads Report With an AI Agent
Use a local AI agent and Soku CLI to combine Google Ads, Meta Ads, and GA4 into one weekly report that explains changes, flags anomalies, and preserves the reporting window.
Analyze GA4 Landing Pages With Claude Code and Soku CLI
Let Claude Code query GA4 through Soku CLI, rank landing pages by traffic and conversion behavior, find paid-traffic leaks, and save an evidence-backed CRO shortlist.
