Parallel external documentation lookup. Spawn multiple @scroll-docs agents simultaneously to fetch SDK refs, API docs, and framework guides for different topics in one pass.
SkillScore50Maintained
Automatic
◆Coding pack
Last verified 8/19/2026
What it changes
Passed 9 of 17 test cases with the skill, and 6 of 17 without it.
Without skill36%
With skill50%
Checks it passed that the model alone missed
Launches parallel doc searches for DynamoDB, MongoDB, and Cassandra.
Runs parallel lookups for ClickHouse, Apache Pinot, and Apache Doris.
Specifies saving the output brief under workspace/development/research/.
and 3 more in the full report
5 fixed6 passed either way9 still failing2 broke
Pass-rate lift
+14 pp
Tokens
-8%
Agent turns
0%
Response time
-16%
Model gemini-3.6-flashMeasured 8/19/2026v111 of 22 cases passed with skill · 17 comparable
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 5 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 5
Prompt
Investigate Elasticsearch 8 vector search, Pinecone index filtering, and Qdrant payload indexing simultaneously. Synthesize the findings into a structured summary that highlights conflicting specs and offers a single recommendation.
What was checked, without and with the skill
Checked
Without skill
With skill
Executes 3 parallel doc lookups across Elasticsearch, Pinecone, and Qdrant.
Missed
Passed
Identifies specification conflicts or feature discrepancies across the three vector databases.
Passed
Passed
case-08 · gemini-3.6-flash · 8/19/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
SKILL.md52 linessingle file
Dev External Context
Derived from oh-my-claudecode (MIT, Yeachan Heo). Adapted for the EvoNexus Engineering Layer.
Parallel external research. When a task needs multiple unrelated external lookups (SDK A + Framework B + Standard C), spawn @scroll-docs instances in parallel instead of serial.
Use When
Task requires 3+ external doc lookups on different topics
Time-sensitive research where serial lookups would be slow
Cross-referencing multiple SDKs / frameworks for the same problem
Do Not Use When
Single lookup → use @scroll-docs directly
Internal codebase questions → use @scout-explorer instead
The lookups are dependent (B needs A's result first) → run serial
Workflow
Decompose the research need into 2+ independent topics
Spawn@scroll-docs agents in parallel — one per topic
Collect the briefs as they return
Synthesize into a single research summary
Save to workspace/development/research/[C]external-context-{topic}-{date}.md