task-summarization
Skill41 starsChanged 6 days agoApache-2.0View on GitHub
Generate concise task summaries with status, timeline, blockers, SLA risk, and recommended next actions
We split this skill into the rules it gives an agent
8 rules. About 0 are common, 5 are distinctive, and 3 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈5
Distinctive · estimate
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 36
Use this when teams need fast context transfer without reading full activity streams.
- Very specific: line 41
Access: Read access to task, task_sla, and sys_journal_field
- Very specific: line 88
If has_breached=true -> include breach reason and recovery action
- Very specific: line 89
If business_percentage>=80 -> mark as urgent follow-up
- Distinctive: line 129
Always include timestamp and summary author context
- Distinctive: line 130
Keep summaries factual; separate facts from recommendations
- Distinctive: line 131
Call out unresolved dependencies explicitly
- Distinctive: line 132
Use consistent summary sections for easier comparison across tasks
≈ “Common” is an estimate: at least 1 in 5 of the 308,232 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Task Summarization
Overview
This skill produces consistent task summaries for handoff, approvals, and management reviews in ServiceNow®. Each summary includes:
- Current status and owner
- Timeline of key events
- Latest customer/agent activity
- SLA exposure and urgency
- Open blockers and next actions
Use this when teams need fast context transfer without reading full activity streams.
Prerequisites
- Roles:
itil,task_admin, oradmin - Access: Read access to
task,task_sla, andsys_journal_field - Input: Target task number or sys_id and intended audience (resolver, approver, manager)
Procedure
Step 1: Read Core Task Context
Retrieve the primary task fields required for summary framing.
Using MCP:
Tool: SN-Get-Record
Parameters:
table_name: task
sys_id: [task_sys_id]
fields: number,short_description,state,priority,assignment_group,assigned_to,opened_at,due_date,sys_updated_onStep 2: Pull Recent Activity and Work Notes
Collect the latest journal updates to build an accurate timeline.
Using MCP:
Tool: SN-Query-Table
Parameters:
table_name: sys_journal_field
query: name=task^element_id=[task_sys_id]^ORDERBYDESCsys_created_on
fields: element,value,sys_created_on,sys_created_by
limit: 30Step 3: Evaluate SLA Risk
Determine if the task is at risk of breach or already breached.
Using MCP:
Tool: SN-Query-Table
Parameters:
table_name: task_sla
query: task=[task_sys_id]^stage=in_progress
fields: sla,planned_end_time,business_percentage,has_breached
limit: 10Decision points:
- If
has_breached=true-> include breach reason and recovery action - If
business_percentage>=80-> mark as urgent follow-up - Otherwise -> keep as standard monitoring
Step 4: Generate Audience-Specific Summary
Tailor summary depth by audience:
- Resolver: technical updates, blockers, dependencies
- Approver: business impact, decision needed, risk if delayed
- Manager: status trend, SLA risk, ownership alignment
Step 5: Save Summary to Work Notes
Store output where downstream stakeholders can reuse it.
Using MCP:
Tool: SN-Add-Work-Notes
Parameters:
table_name: task
sys_id: [task_sys_id]
work_notes: |
TASK SUMMARY
- Status:
- Latest update:
- SLA risk:
- Blockers:
- Next action:Tool Usage
| Tool | Purpose |
SN-Get-Record | Get canonical task context |
SN-Query-Table | Retrieve journals and SLA details |
SN-Execute-Background-Script | Build aggregated timelines for complex tasks |
SN-Add-Work-Notes | Persist summary in task history |
Best Practices
- Always include timestamp and summary author context
- Keep summaries factual; separate facts from recommendations
- Call out unresolved dependencies explicitly
- Use consistent summary sections for easier comparison across tasks
- Re-summarize after major state or owner transitions
Troubleshooting
Summary Misses Critical Context
Symptom: Key issue details are absent from the generated summary.
Cause: Journal query window is too short or excludes relevant elements.
Solution: Increase journal limit and include both comments and work notes.
Conflicting Status Signals
Symptom: Task state and SLA risk appear contradictory.
Cause: State updated without corresponding SLA recalculation yet.
Solution: Re-query task_sla and compare with latest task update timestamp.
Related Skills
admin/task-analysis- Analyze task trends and bottlenecksitsm/incident-activity-summarization- Summarize incident-specific activity streamscatalog/approval-summarization- Summaries tailored for approval workflows
References
- https://github.com/ServiceNow/ServiceNowDocs/blob/australia/markdown/it-service-management/now-assist-for-it-service-management-itsm/now-assist-itsm-skills.md
- https://github.com/ServiceNow/ServiceNowDocs/blob/australia/markdown/field-service-management/now-assist-for-field-service-management-fsm/cust-now-assist-fsm-wot-summarization-skill.md
In this repository184 other files it gives its agents.
- happy-platform-skillsSkill
- application-scopeSkill
- batch-operationsSkill
- configuration-validationSkill
- deployment-workflowSkill
- generic-crud-operationsSkill
- instance-managementSkill
- schema-discoverySkill
- script-executionSkill
- script-syncSkill
- smart-assessmentSkill
- task-analysisSkill
- update-set-managementSkill
- user-provisioningSkill
- workflow-creationSkill
- workspace-insightsSkill
- approval-summarizationSkill
- approval-workflowsSkill
- catalog-builder-aiSkill
- catalog-item-generationSkill
- item-creationSkill
- multi-turn-orderingSkill
- request-fulfillmentSkill
- ui-policiesSkill
- variable-managementSkill
- ci-discoverySkill
- cmdb-search-analysisSkill
- data-qualitySkill
- impact-analysisSkill
- relationship-mappingSkill
- service-graph-diagnosisSkill
- activity-responseSkill
- csm-case-summarizationSkill
- chat-recommendationSkill
- email-recommendationSkill
- csm-kb-generationSkill
- resolution-notesSkill
- csm-sentiment-analysisSkill
- csm-sidebar-summarizationSkill
- csm-suggested-stepsSkill
- trending-topicsSkill
- automated-testingSkill
- business-rulesSkill
- client-scriptsSkill
- code-assistSkill
- code-reviewSkill
- data-importSkill
- debugging-techniquesSkill
- fluent-sdkSkill
- glide-api-referenceSkill
- mcp-server-installationSkill
- mcp-serverSkill
- notificationsSkill
- performance-optimizationSkill
- scheduled-jobsSkill
- scripted-rest-apisSkill
- script-includesSkill
- servicenow-docs-mcpSkill
- test-generationSkill
- ui-actionsSkill
- The first 60 shown.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
Reviews can't be read right now.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

