a452cd95b8
- Every workflow prompt now shows a plan and asks the user to confirm before executing - /autoresearch asks for execution environment (local, branch, venv, cloud) and confirms before looping - Writer agent and key prompts now generate charts (pi-charts) and diagrams (Mermaid) when data calls for it - Cite alphaXiv and Agent Computer in README and website homepage - Clear terminal screen before launching Pi TUI - Remove Alpha Hub GitHub link in favor of alphaxiv.org Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
16 lines
856 B
Markdown
16 lines
856 B
Markdown
---
|
|
description: Compare a paper's claims against its public codebase and identify mismatches, omissions, and reproducibility risks.
|
|
args: <item>
|
|
section: Research Workflows
|
|
topLevelCli: true
|
|
---
|
|
Audit the paper and codebase for: $@
|
|
|
|
Requirements:
|
|
- Before starting, outline the audit plan: which paper, which repo, which claims to check. Present the plan to the user and confirm before proceeding.
|
|
- Use the `researcher` subagent for evidence gathering and the `verifier` subagent to verify sources and add inline citations when the audit is non-trivial.
|
|
- Compare claimed methods, defaults, metrics, and data handling against the actual code.
|
|
- Call out missing code, mismatches, ambiguous defaults, and reproduction risks.
|
|
- Save exactly one audit artifact to `outputs/` as markdown.
|
|
- End with a `Sources` section containing paper and repository URLs.
|