trace

Search git history, repos, and Oracle memory to locate project files and code.

1|Updated May 24, 2026
One-click install
npx skills add https://github.com/tinner-deinno/innova-skills-lib --skill trace-tinner-deinno
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: trace
Source: https://github.com/tinner-deinno/innova-skills-lib/tree/main/core/trace
Command: npx skills add https://github.com/tinner-deinno/innova-skills-lib --skill trace-tinner-deinno

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

trace helps you quickly find the right project files, code, and related knowledge even when you don’t know where it lives or when it was last changed.

Core Features & Use Cases

  • Unified discovery across history and memory: searches current repo contents plus Oracle memory traces and prior knowledge/logs to surface where something exists.
  • Depth options for hard-to-find items: runs in fast Oracle-only mode, a smart Oracle-then-escalate mode, or a wave-based deep search across files, git history, cross-repo matches, and GitHub issues/PRs.
  • Distills results into a persistent trace log: writes a dated trace record with friction scoring and goal-backward confidence checks so future searches get faster and more reliable.

Real-world use case: you remember a project feature name or config key but can’t recall which repo or file it’s in; trace can search your codebase, git history, and Oracle memory and then log the findings for later reuse.

Quick Start

Ask for a trace of what contains the keyword by running: /trace <query>

Frequently Asked Questions about trace

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

FAQPage Schema
How do I search across git history and multiple repos to find where a specific code feature is located?

You can locate project files across git history and multiple repos by running a unified discovery query that searches current codebases, commit logs, and memory traces. This surfaces exact file locations and prior project context even when the target repository is uncertain.

What is the best way to find a project config key when I don't know which repository it lives in?

Finding a project config key across uncertain repositories requires a cross-repo search tool that queries current files and git history simultaneously. Results are distilled into a persistent trace log with friction scoring and confidence checks to accelerate future lookups.

How do I run a deep search across GitHub issues and PRs for cross-codebase troubleshooting?

To run a deep search across GitHub issues and PRs for cross-codebase troubleshooting, initiate a wave-based deep search mode. This escalates beyond current files to scan commit history, cross-repo matches, and issue threads to locate uncertain project targets.

Can I use trace logging to improve code discovery accuracy for future searches?

Yes, trace logging improves code discovery accuracy for future searches by writing a dated trace record after each goal-backward confidence check. These persistent logs capture prior knowledge and friction scores, making subsequent project tracing faster and more reliable.

Do I need any specific dependencies to perform session mining on old git history?

No specific dependencies are required to perform session mining on old git history. You can execute optional dig session mining via command flags directly within the discovery tool to extract prior context and historical project traces.

When should I use smart Oracle-then-escalate mode versus fast Oracle-only mode for project tracing?

Use fast Oracle-only mode for quick project tracing lookups, and use smart Oracle-then-escalate mode when initial memory traces are insufficient. The smart mode escalates to wave-based deep search across files and git history if the target location remains uncertain.