Autonomous agent for research and analysis tasks. Trigger with "investigate with Task", "research this", "analyze this", "search for", "gather information", "run in background".
NewMaintained
Automatic
◆Research pack
Last verified 8/15/2026
What it changes
Passed 6 of 18 test cases with the skill, and 12 of 18 without it.
Without skill67%
With skill33%
Checks it passed that the model alone missed
Recommends using deep reasoning commands like ultrathink rather than Task.
Notes that Task is meant for gathering external or multi-file context rather than internal theoretical trade-off analysis.
Explains that architectural implementation design requires approval planning rather than multi-source information collection.
2 fixed5 passed either way11 still failing4 broke
Pass-rate lift
-33 pp
Tokens
+25%
Agent turns
0%
Response time
-21%
Model gemini-3.6-flashMeasured 8/15/2026v17 of 22 cases passed with skill · 18 comparable
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 2 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 2
Prompt
We are evaluating the architectural trade-offs of monolithic versus microservices architectures using pure theoretical reasoning from your existing training data. Should we run Task or ultrathink?
What was checked, without and with the skill
Checked
Without skill
With skill
Recommends using deep reasoning commands like ultrathink rather than Task.
Missed
Passed
Notes that Task is meant for gathering external or multi-file context rather than internal theoretical trade-off analysis.
Missed
Passed
case-06 · gemini-3.6-flash · 8/15/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
SKILL.md223 linessingle file
Autonomous agent for research and analysis tasks
Launches a smart agent to handle complex searches and investigations. Great for large-scale work without eating up context.
Usage
# Request Task from Claude
"Investigate [task] using Task"
What Task Does
Works Independently
Combines multiple tools automatically
Gathers and analyzes step by step
Puts results together in clear reports
Saves Context
Uses less memory than manual searching
Searches lots of files efficiently
Pulls data from outside sources
Ensures Quality
Checks if sources are reliable
Verifies from different angles
Fills in missing pieces
Basic Examples
# Complex codebase investigation
"Investigate which files implement this feature using Task"
# Large-scale file search
"Identify configuration file inconsistencies using Task"