Plugin Claude Code
Lateral-thinking skill for AI coding agents. Breaks debugging tunnel vision and surfaces goal-appropriate alternatives across Grounded, Balanced, and Wild tiers.
Plugin Claude Code
Lateral-thinking skill for AI coding agents. Breaks debugging tunnel vision by surfacing end goals, solution horizons, and goal-appropriate alternatives across three tiers: Grounded, Balanced, and Wild.
Skill Claude CodeCodex
Structured brainstorming skill for features, architecture, and ideas. Activates automatically during plan mode. Generates ideas in three tiers: Grounded (practical, shippable), Balanced (assumption-questioning, loosely feasible), Wild (speculative, constraint-free). Use when user invokes /lateralus-brainstorm, says…
Skill Claude CodeCodex
Lateral-thinking escape hatch for stalled debugging. Surfaces end goal and solution horizon, generates alternatives in three tiers: Grounded (testable causes), Balanced (assumption-questioning), Wild (speculative reframes), plus Workaround (bypass now, fix later). Use when user invokes /lateralus, says "stuck", "going…
Instructions file CodexOpenCode
Instructions for SunnyKTuladhar/lateralus, covering lateralus — project context, project overview, what lives where, file ownership — edit only these and ci sync workflow.
Instructions file
Instructions for SunnyKTuladhar/lateralus, a project described as: 🌀The Lateral Thinking skill, for when you need to turn you coding agent to brainstorm, solve issues or to overthink and overanalyse stuff. Three thinking levels (grounded, balanced and wild) to think to suit your required needs.
Instructions file Gemini CLI
Instructions for SunnyKTuladhar/lateralus, a project described as: 🌀The Lateral Thinking skill, for when you need to turn you coding agent to brainstorm, solve issues or to overthink and overanalyse stuff. Three thinking levels (grounded, balanced and wild) to think to suit your required needs.
Agent
Tier 2 middle-ground ideator. Sits between lateralus-ideator-ground (Tier 1, purely testable) and lateralus-ideator-wild (Tier 3, purely speculative). Generates 3-5 hypotheses that question assumptions while remaining loosely verifiable. Use for a single-pass ideation when running all tiers separately is overkill.
Agent
Tier 1 grounded ideator. Takes the context block from the lateralus skill interrogation phase and generates 3-5 concrete, testable hypotheses outside the obvious failure category. Horizon-calibrated depth. Never implements. Pair with lateralus-ideator-wild for full coverage.
Agent
Tier 3 wild-reframe generator. Takes the context block from the lateralus skill interrogation phase and generates 3-5 speculative reframes designed to jolt new associations and break tunnel vision. Not literal fixes. Never implements. Pair with lateralus-ideator-ground (Tier 1).
Agent
Makeshift bypass generator. Gets the user unblocked without fixing root cause. Use when horizon is demo, deadline, POC, test, or just-need-to-move. Always produces a debt log. Never dresses up a bypass as a real fix.