replicate-paper
OfficialReplicate and extend published results end-to-end.
Education & Research#experiment planning#research automation#PDF parsing#mechanistic interpretability#paper replication#causal models
Authorgoodfire-ai
Version1.0.0
Installs0
System Documentation
What problem does it solve?
It helps you reproduce a paper’s claimed experiments and compare the outcomes against the original methodology so you can validate, debug, or extend the research rather than re-implementing everything from scratch.
Core Features & Use Cases
- Paper-to-replication orchestration: Extracts methodology and enumerates the paper’s experiment list, then routes you through the standard pipeline for planning, execution, and interpretation.
- Paper-grounded comparison artifacts: Builds paper-specific sidecar context (including reported quantitative findings) so the final report includes a clear paper-vs-replication comparison.
- Session-safe workflow: Enforces the research-session conventions (active session required; one session, one plan) and supports interactive approval gates for extracted methodology and generated drafts.
- Use cases: Replicating a published causal-mechanistic study, validating benchmark claims, or extending results with additional experiments using the same experimental skeleton.
Quick Start
Run replicate-paper with the paper file path, for example: Ask your agent to execute the /replicate-paper skill using the argument "/path/to/paper.pdf".
Dependency Matrix
Required Modules
None requiredComponents
Standard package💻 Claude Code Installation
Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.
Please help me install this Skill: Name: replicate-paper Download link: https://github.com/goodfire-ai/causalab/archive/main.zip#replicate-paper Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
Agent Skills Search Helper
Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.