just-vibeGeneral · inspect by default
/challenge
Identify weak assumptions, complexity, and failure cases
Use to test a proposal's assumptions, not to manufacture objections; decision-premortem explores a future failure story.
Make it your own.
In Claude Code, use the slash command and add your context. In Codex, select challenge from the just-vibe skill picker, then send the same brief.
Version 0.11.0 also supports /jv challenge, /just-vibe challenge and /jv:challenge in Claude. See shortcut setup and context examples.
/just-vibe:challenge Challenge the assumption that every notification needs a separate service./just-vibe:challenge Challenge a cache proposal whose stale data could affect authorization./just-vibe:challenge Challenge this proposal without traffic data; do not invent capacity limits.What the agent does
- Extract the critical assumptions as falsifiable claims.
- Seek disconfirming evidence and plausible edge cases in current code, workload and constraints.
- Rank issues by impact and likelihood, distinguishing observed defects from speculative risks.
Inputs
- proposal, plan, architecture, or hypothesis and its success criteria.
Optional context: scope, references, constraints, successCriteria, environment, mode, budget.
Scope
- Reads
- Test assumptions and failure modes; no adversarial criticism for its own sake and no implementation changes.
- Writes
- No source changes in inspect/plan. Save only requested planning artifacts. A separately requested repair uses the relevant implementation workflow.
- Mode
- Inspect; proposal, plan, architecture, or hypothesis and its success criteria.
- Prerequisites
- Resolve the user brief and inspect the relevant project or supplied evidence. External capabilities are optional unless the selected action actually needs them.
Expected output
- Ranked challenges, each with its trigger, consequence, evidence and a discriminating experiment, plus smaller or more robust alternatives where they exist.
How the work is checked
- A supported assumption remains accepted; a serious unsupported dependency receives a concrete validation step.
When to stop or clarify
- Separate demonstrated flaws from speculative concerns. Do not demand certainty where a reversible experiment resolves the risk.
Handling missing context
- Infer
- Resolve the named files, existing scripts, current task and earlier corrections from the conversation and repository.
- Assume
- Use the narrowest interpretation that completes a reversible local task; state a consequential assumption once.
- Ask
- Ask when competing targets or incompatible success conditions would change the result; continue independent inspection first.
Technical guidance
- Evidence
- Inspect the proposal's claimed benefits, assumptions, constraints and failure/recovery model.
- Method
- Construct concrete counterexamples and identify evidence that could falsify the key assumption.
- Pitfall
- Contrarian preferences or speculative catastrophes without a mechanism add noise.
- Check
- Test whether the proposal survives the strongest relevant counterexample and distinguish resolved from open objections.
Situational decisions
When an assumption survives the available evidence: Keep it accepted provisionally and name evidence that would reverse the conclusion.
The coding agent follows this workflow using its available tools. Installation does not grant service access or guarantee an outcome. Read the compatibility notes.