TikTok Army
Agent fleet · Studio-powered
Workflow
OverviewNew briefWorkflowsApprovals
Operations
Agent catalogAgent runsRun indexAlerts
Domain
PostsBrandsCompetitors
Knowledge
User guideBuilder docsSkills
System
Settings
Read-only.
Writes route through agents → spend cap → compliance.
envproduction
read-only · writes route through agents
Updated just now
← synthesis

Run 3c39a091…

3c39a091-4950-49f5-87a6-bf67266b6baf

succeeded
Agent
synthesis
Trigger
manual
Latency
190ms
Cost
$0.0114
Model
claude-sonnet-4-6
Brand
00000000…
Started
2026-04-30 17:11:02
Completed
2026-04-30 17:11:02
Input4 lines
{
  "workflow_slug": "profile_audit",
  "user_message_preview": "Workflow: Profile Audit (profile_audit)\nBrief: handle='lakucosmetics', target_type='third_party'\n\nUpstream outputs:\n\n## audit_health\n```json\n{\n  \"checked_count\": 7,\n  \"alerts_raised\": 2,\n  \"alerts_deduped\": 0,\n  \"per_account\": [\n    {\n      \"account_id\": \"3478a0fc-ad04-4f8d-9f00-c06b9e4477cb\",\n      \"handle\": \"cleanbeauty\",\n      \"checks_run\": [\n        \"provider_status\",\n        \"engagement_drop\",\n        \"revenue_drop\",\n        \"rate_limit\"\n      ],\n      \"alerts_raised\": 0,\n      \"alerts_dedu"
}
Output4 lines
{
  "cost_usd": 0.011388299999999999,
  "report_md": "# Profile Audit — Mock Report\n\n## Summary\nAccount health is **72/100** — engagement is trending down 3% week-over-week, and posting cadence (2/wk) is below the niche median (4–5/wk). No shadowban signal, no policy flags. Compliance pass with one FTC substantiation warning.\n\n## Top three actions\n1. **Double posting cadence** — push to 4 posts/week, weighted to lo-fi POV formats.\n2. **Pivot creative** — the shelf-reveal POV format is over-indexing on saves vs studio-styled hero shots. Brief Studio for 3 more in this register.\n3. **Substantiate the '24h hydration' claim** in caption metadata before re-running paid.\n\n## Audience opportunity\nThree segments mapped: *Morning Routine Minimalists* (medium, 0.84 fit), *Glow-curious Teens* (large, 0.71), *Clinical-skin Researchers* (small, 0.62). Recommend Spark prospecting against Teens, Shop conversion against Researchers.\n\n## Competitive intel\nCrawled 10 competitors over the last 30 days. **2 viral outliers** (both `problem_solution` hooks; 9.2× and 6.4× their creator medians). Hashtag gap: competitors lean heavily on `#minimaldewy` and `#dermtested` — we use neither. Posting heatmap clusters at 7–9 PM UTC Tue/Wed/Thu.\n\n## Trends to lean into next 14 days\n- `#cleanbeauty` (+38% lift), `#minimaldewy` (+27%)\n- Sound: *Espresso (sped up)* — high fit\n- Format: lo-fi shelf reveal, no overlay text\n\n*This is a mock report generated without a live Claude API key. Set `CLAUDE_MODE=real` and provide `ANTHROPIC_API_KEY` to run for real.*\n"
}

Started 2d ago · completed 2d ago