AgentStack
SKILL verified MIT Self-run

Devils Advocate

skill-saptarsibhowmick-claude-workflow-for-econ-phd-devils-advocate · by saptarsibhowmick

Structured debate between two perspectives on a research design decision.

No reviews yet
0 installs
9 views
0.0% view→install

Install

$ agentstack add skill-saptarsibhowmick-claude-workflow-for-econ-phd-devils-advocate

✓ scanned · ✓ verified — works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access No
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

Are you the author of Devils Advocate? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

Devil's Advocate — Structured Design Debate

When to Use

When uncertain about a design choice: DiD vs synthetic control, instrument A vs instrument B, cluster at district vs province, include control X or not.

Steps

  1. State the decision clearly: what is being decided, what are the options
  2. Argue FOR Option A — the strongest possible case:
  • What assumptions does it require?
  • What makes it the better choice for this specific setting?
  • What would a supportive referee say?
  • What evidence supports it?
  1. Argue FOR Option B — the strongest possible case:
  • Same structure as above
  • No straw-manning — the best version of this argument
  1. Present the trade-off matrix:

| Dimension | Option A | Option B | |---|---|---| | Assumptions required | | | | Robustness to violations | | | | Data requirements | | | | Reviewer acceptability | | | | Precision | | |

  1. Recommendation (with caveats): which option is stronger and why
  2. Document the decision and reasoning in pre-analysis-plan/strategy-memo.md

Important

  • Ideally run with two separate Claude sessions (one per side) for genuine independence
  • The goal is to surface the strongest version of each argument, not to confirm a prior
  • Document the decision — future you will want to know why you chose Option A

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet — be the first.

Versions

  • v0.1.0 Imported from the upstream source.