| name | second-pass-grasp |
| description | Full-text deep-read of a paper's Introduction, Method, and Results sections to draft the structured bundle (problem, method, key results with hedge levels, limitations), each field carrying a precise source anchor. Use this after first-pass-skim has classified the paper and drafted candidate angles — this is the main content-extraction pass of the reading pipeline. |
| execution | subagent |
| prompt | ./prompt.md |
| input | paper_ref (string), paper_type (string), skim_notes (string) |
| output | draft_bundle (dict), uncertain_fields (list of strings) |
| dependencies | {"sops":["spawn-agent","literature-research"]} |
Second Pass Grasp
Keshav's second pass: full read of Introduction/Method/Results, drafting the structured bundle with precise source anchors and per-claim hedge levels.
Execution
Subagent — spawned via subagent-spawning/spawn-agent skill.
Why Subagent
Full-text extraction across multiple sections into a consistent schema benefits from a dedicated context — avoids extraction drift between fields that a shared/shorter context could introduce.
Available SOPs
Optional, no fixed order; the final leaf is always a sop.
| SOP | When to use |
|---|
| spawn-agent | Spawn a customized CC subagent with full MCP tool access. |
| literature-research | Full-text paper reading via alphaxiv. Import from literature-engine. |
