Select a read-only OpenCode workflow based on question uncertainty and framing.
npx skills add https://github.com/vhqtvn/vh-solara --skill think-mode
Please help me install this Agent Skill. Skill: think-mode Source: https://github.com/vhqtvn/vh-solara/tree/main/.opencode/skills/think-mode Command: npx skills add https://github.com/vhqtvn/vh-solara --skill think-mode
Think-mode helps you decide which read-only OpenCode workflow fits a question before you spend time on the wrong shape of analysis. It prevents unnecessary research, debate, or solution-brief work when the framing is already clear or when the uncertainty is still evolving.
Ask the think-mode skill which read-only workflow should handle your question and provide the current context, uncertainty, and desired outcome.
High-intent search queries and answers about installing and using this skill.
Choosing a read-only workflow requires evaluating your question's uncertainty to select between researcher, debate, or solution-brief modes, preventing wasted analysis on the wrong session shape. The skill returns a scaffolded prompt with operator gating to guide the process.
Use a debate workflow when distinguishing factual gaps from active choice-making, and use a solution-brief when stable framing is needed. The skill evaluates stakeholder-driven ambiguity to determine whether one-shot or phased solution-briefing is appropriate.
The best way to select an OpenCode workflow is by assessing whether facts, options, or framing need different treatment. The skill applies uncertainty shaping to choose among direct build, researcher, debate, and solution-brief workflows.
Yes, you need to re-evaluate workflow selection when a new stakeholder constraint means the plan should be reconsidered. The skill distinguishes stable framing from stakeholder-driven ambiguity to prevent unnecessary research or debate on outdated parameters.
To start workflow selection, ask the skill which read-only workflow should handle your question and provide the current context, uncertainty, and desired outcome. It then returns a scaffolded prompt with operator gating for execution.
dalawwaLaunch a sub-agent judge to evaluate conversation results with evidence-based scoring.
hongbietcodeFacilitate multi-agent debate across Researcher, Critic, and Synthesizer roles.
plurigridPerforms causal inference with structural models and Python libraries like DoWhy, CausalML, and Pyro.
product-on-purposeGenerate a Jobs to be Done canvas with structured customer job sections.
