devils-advocate

Plugin devils-advocate from marketplace stellarshenson-marketplace in stellarshenson/claude-code-plugins. Commands read from the repo's manifest — 3 days ago.

Skills (6)

• adversarial-review

  Hostile, independent review - spawn fresh context-free reviewer subagents that try to BREAK a change. Two modes - diff bug-hunt and whole-repo audit. Multi-round - find, fix, re-confirm. Use before a risky commit/merge, a UI or terminal-UI ship, a shell installer, trusting an experiment's verdicts o

• evaluate

  Generate baseline devil's advocate evaluation. Produces concern catalogue with risk scores and scorecard. Requires setup to have been run first (devils_advocate.md and fact_repository.md must exist).

• improve

  Bridge between evaluate and iterate. Asks user how to address devil's concerns - user suggestions, auto-apply recommended options, or planning mode. Now integrated into iterate's Step 1 - use /devils-advocate:iterate directly.

• iterate

  Improve document based on devil's concerns, create versioned file, re-score, rename with residual. Clear 4-step workflow - improve, version, score, rename. Re-scores in place when user made changes outside Claude.

• run

  Run the full devil's advocate workflow end-to-end. Setup persona, evaluate baseline, then iterate corrections until residual risk is acceptable. Use when the user wants the complete analysis in one go.

• setup

  Set up devil's advocate persona and fact repository for a target document. Use when starting a new critical analysis. Builds the devil persona (role, biases, triggers) and harvests verified facts from source materials.

Slash commands (5)

• /devils-advocate:adversarial-review

  Hostile independent review by spawning fresh reviewer subagents that try to BREAK a change - invokes the devils-advocate:adversarial-review skill, seeding one of the expert adversaries; find, fix, re-confirm clean

• /devils-advocate:evaluate

  Generate the baseline concern catalogue and scorecard from the devil persona - step 2 of the workflow

• /devils-advocate:iterate

  One improvement cycle - decide approach, apply changes, version, re-score, rename with residual - step 3 of the workflow

• /devils-advocate:run

  Run the full devil's advocate critical analysis workflow

• /devils-advocate:setup

  Build the devil's-advocate persona and harvest the fact repository for a target document - step 1 of the workflow

Subagents (2)

• adjudicator

  Turns adversarial-review findings into ONE change plan - materiality triaged first, grouped by root cause, the smallest changes that answer them without seeding the next round, and loop-introduced machinery reverted rather than refined. Use after an adversarial review returns findings: a multi-lens

• adversarial-reviewer

  Hostile red-team reviewer that tries to BREAK a change rather than approve it. The caller names one expert lens - architect, bug-hunter, qa-engineer, analyst, digital-marketer, data-scientist, methodologist, ux-designer, tui, devops, popular-science, slop-hunter or ai-engineer - and the reviewer arg

Similar plugins