trace

Orchestrate parallel tracer hypotheses to identify evidence-based causal explanations.

1|Updated Mar 27, 2026
One-click install
npx skills add https://github.com/oneforce/oh-my-gemini-cli --skill trace-oneforce
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: trace
Source: https://github.com/oneforce/oh-my-gemini-cli/tree/main/skills/trace
Command: npx skills add https://github.com/oneforce/oh-my-gemini-cli --skill trace-oneforce

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Restate ambiguous, causal observations and orchestrate multiple tracer hypotheses in parallel to surface explanations grounded in evidence rather than raw outputs.

Core Features & Use Cases

  • Orchestrates Gemini built-in team mode with a default of three tracer lanes, each pursuing a distinct hypothesis.
  • Enforces a worker contract: restate lane, gather evidence for and against, rank evidence, identify the critical unknown, and propose the best discriminating probe.
  • Supports a structured reasoning framework using an evidence-strength hierarchy and cross-check lenses (systems, premortem, science) for robust analysis.
  • Ideal for runtime bugs, performance mysteries, architectural postmortems, and experimental results requiring multi-source validation.

Quick Start

Summarize the observation and invoke /trace to spawn three parallel tracer lanes in Gemini built-in team mode.

Frequently Asked Questions about trace

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I debug runtime bugs using evidence-driven tracing?

Parallel hypothesis tracing works by spawning three concurrent tracer lanes in Gemini team mode. Each lane restates the observation, gathers evidence for and against a distinct hypothesis, ranks the evidence, and proposes the best discriminating probe.

How do I investigate performance issues with parallel hypotheses?

To investigate performance issues, summarize the observation and invoke the trace command. This spawns three parallel tracer lanes that gather and rank evidence to identify the critical unknown and propose a discriminating probe for robust analysis.

Does Gemini team mode support architectural postmortem analysis?

Hypothesis tracing requires the built-in tracer agent and Gemini team-mode orchestration. It enforces a worker contract where each lane restates observations, gathers evidence for and against, ranks evidence, identifies critical unknowns, and proposes discriminating probes.

When should I use parallel tracer lanes for debugging?

Use parallel tracer lanes for debugging runtime bugs, performance mysteries, architectural postmortems, and experimental results. It is ideal when strong, evidence-based reasoning is required to resolve ambiguous, causal observations rather than relying on raw outputs.