Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →shobcoder/deep-research
GOD MODE deep research skill. Multi-phase autonomous research loop: query decomposition → multi-source crawling → claim cross-referencing → conflict resolution → structured synthesis with inline citations.
Passed 19 of 22 test cases with the skill, and 10 of 22 without it.
and 8 more in the full report
9 fixed10 passed either way3 still failing0 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 9 test cases the skill fixed. Same model, same prompt, with and without the skill.
While gathering market data on nuclear fusion startup funding, no verified figures could be found for a specific privately held firm. What specific placeholder string must be inserted instead of estimating?
| Checked | Without skill | With skill |
|---|---|---|
| Missing numerical data is represented strictly using the text bracket string [no data found] | Missed | Passed |
The instructions your model reads when the skill loads.
/deep-research <query> or when user says "research this deeply", "go deep on",
"full research report on", "investigate this thoroughly".
Raw search results are noise. Verified synthesis is signal. Every claim needs a source. Every conflict needs a resolution. A great deep research report is a structured intelligence brief, not a search summary.
Query
└── Phase 1: Decompose → Sub-questions
└── Phase 2: Parallel Search → Raw Sources
└── Phase 3: Crawl & Extract → Claims
└── Phase 4: Cross-Reference → Verify / Conflict
└── Phase 5: Synthesize → Report
└── Phase 6: Quality Gates → Deliver
Break the user's query into 3–7 atomic sub-questions. Each must be:
Example:
Query: "Is Company X profitable?"
Sub-questions:
test cases passed, vs 10 / 22 without
Published skills whose descriptions closely match this one, with their headline benchmark lift.
More skills from the same source and nearby in the registry, with their headline benchmark lift.