Scheduled Tasks — 15 Automated Task Templates

Scheduled tasks are recurring workflows that execute on a defined schedule. Each task is a structured template containing the skills to invoke, the data to pull, the output format to follow, and the approval tier that governs what happens with the result.

Task Categories

Monitoring Tasks (M-series)

Task	Schedule	Tier	Skills Used
m1-rankings-monitor	Weekly — Monday 7 AM	Autonomous	geogrid-analysis, localseodata-tool
m2-review-velocity	Daily — 8 AM	Autonomous	review-management, localseodata-tool
m3-gbp-change-monitor	Daily — 6 AM	Autonomous	gbp-optimization, localseodata-tool
m4-lsa-rankings-monitor	Weekly — Monday 7 AM	Autonomous	lsa-ads, localseodata-tool
m5-ai-visibility-monitor	Monthly — 1st, 9 AM	Autonomous	ai-local-search, localseodata-tool

Reporting Tasks (R-series)

Task	Schedule	Tier	Skills Used
r1-weekly-report	Weekly — Friday 5 PM	Autonomous	local-reporting, localseodata-tool
r2-monthly-client-report	Monthly — 1st, 10 AM	Queue	local-reporting, client-deliverables
r3-multi-location-rollup	Monthly — 2nd, 8 AM	Queue	multi-location-seo, local-reporting
r4-quarterly-business-review	Quarterly — 1st of quarter	Queue	local-reporting, client-deliverables

Engagement Tasks (E-series)

Task	Schedule	Tier	Skills Used
e1-gbp-post-drafts	Monthly — 1st, 8 AM	Queue	gbp-posts, localseodata-tool
e2-review-response-drafts	Daily — 9 AM	Queue	review-management, localseodata-tool
e3-citation-audit	Monthly — 15th, 8 AM	Autonomous	local-citations, localseodata-tool
e4-page-content-audit	Quarterly — 1st of quarter	Autonomous	local-landing-pages, localseodata-tool

Prospecting Tasks (P-series)

Task	Schedule	Tier	Skills Used
p1-prospect-audit	On-demand	Autonomous	local-seo-audit, localseodata-tool
p2-competitor-monitor	Monthly — 1st, 7 AM	Autonomous	local-competitor-analysis, localseodata-tool

TASK.md Format

Every task is defined in a TASK.md file within its directory. The file contains:

Metadata Header

name: m1-rankings-monitor
description: Weekly geogrid ranking scan. Tracks ARP, ATRP, and SoLV
  week-over-week. Alerts when rankings degrade.
schedule: weekly — Monday 7 AM
tier: autonomous
skills: geogrid-analysis, localseodata-tool
mcps: LocalSEOData

Skills Section

Lists primary skills required and fallback guidance if a skill or MCP connection is unavailable. Includes key metric definitions and interpretation rules so the task can produce meaningful output even without live data.

Verification Checklist

Pre-execution confirmations the agent checks before running:

1. Required skills available
2. Location brief exists
3. MCP connections responsive
4. Prior scan data accessible (for comparison)
5. Output directory exists

Prompt Template

The execution instructions with variable placeholders:

Variables: {BUSINESS_NAME}, {LOCATION}, {PRIMARY_KEYWORD},
           {RADIUS_MILES}, {TODAY}

The prompt template guides the agent through:

• Data collection (which endpoints to call)
• Analysis (what to compare, what thresholds apply)
• Output generation (where to write, what format to follow)
• Alert evaluation (when to trigger notifications)

Output Configuration

Output paths:
  Primary: scans/{date}-geogrid.md
  Alert (conditional): alerts/{date}-ranking-drop.md

Every task specifies exactly where its output lands in the brief directory structure.

Alert Thresholds

Default sensitivity levels with override capability:

ARP degradation: 2+ positions → alert
SoLV drop: 10+ points → alert
Configurable via brand brief

Task Execution Flow

1. Schedule triggers — task fires at configured time
2. Verification — agent checks prerequisites
3. Execution — skills and tools run according to prompt template
4. Output — results written to brief directory following output schema
5. Brief update — session log entry added, findings updated if needed
6. Tier evaluation — output routed based on approval tier:
   • Autonomous: complete, optional notification
   • Queue: draft held, approval request sent
   • Notify: pre-approval sent, awaits confirmation
7. Alert check — thresholds evaluated, alerts fired if crossed

Status Handling

Task outputs include a status field:

• SUCCESS — all tools executed, complete data returned
• PARTIAL — some tools failed; output generated from available data, errors logged
• FAILED — task incomplete; errors logged, brief flagged for review

The system does not silently skip failures. Every problem is documented and flagged for human review.

Customizing Task Schedules

Override default schedules per client in the brand brief:

• Adjust frequency (weekly → daily, monthly → biweekly)
• Shift timing to match client timezone or reporting windows
• Disable tasks not relevant to a specific client
• Change approval tiers based on client trust level

Output File Naming

All task outputs follow: briefs/{brand}/{location}/{category}/{YYYY-MM-DD}-{task-type}.md

Examples:

briefs/keystone-insurance/buffalo/reports/2026-04-01-weekly.md
briefs/keystone-insurance/buffalo/scans/2026-04-07-geogrid.md
briefs/keystone-insurance/buffalo/drafts/2026-04-01-gbp-posts.md
briefs/keystone-insurance/buffalo/alerts/2026-04-07-ranking-drop.md
briefs/keystone-insurance/buffalo/prospects/2026-04-04-prospect-audit.md

Output Categories

Category	Purpose
reports/	Weekly, monthly, quarterly summaries
scans/	Geogrid scans, citation audits, competitor snapshots
drafts/	GBP posts, review responses, page content awaiting approval
alerts/	Monitoring alerts requiring immediate attention
prospects/	Prospect audits for sales use

---

Learn More

To learn what scheduled tasks can do for your local SEO workflow, see the scheduled tasks overview.