trace

Find projects across git history, repositories, docs, and Oracle memory.

Updated Feb 28, 2026
One-click install
npx skills add https://github.com/heman1033p-create/cv-v2-kongphop --skill trace-heman1033p-create
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: trace
Source: https://github.com/heman1033p-create/cv-v2-kongphop/tree/main/.gemini/skills/trace
Command: npx skills add https://github.com/heman1033p-create/cv-v2-kongphop --skill trace-heman1033p-create

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve? Locating a specific project, file, or commit scattered across local repos, git history, GitHub issues, and past session memory is slow and error-prone. This Skill unifies those search surfaces into a single trace command with tiered depth. ## Core Features & Use Cases - Tiered Search Modes: Choose --oracle for fast memory-only lookups, --smart for Oracle-first search with automatic escalation, or --deep for five parallel search agents covering files, git history, GitHub issues, other repos, and memory. - Flexible Repo Targeting: Search the current repo by default, point at a local path with --repo, or clone a remote URL into ghq before searching. - Persistent Trace Logs: Every search writes a timestamped markdown log to ψ/memory/traces/ and records results to Oracle, so future searches benefit from past findings. - Use Case: You remember building a rate-limiter somewhere last month but not where. Run a trace query and get matching files, commits, issues, and prior session notes compiled into one log. ## Quick Start Ask the AI to trace where the rate limiter project lives across my repos and git history.

Frequently Asked Questions about trace

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

FAQPage Schema
How do I find a project across multiple git repositories?▼

Run a trace query with the project name and the skill searches the current repo, git history, GitHub issues, other repos under ghq, and Oracle memory. Use --deep mode to launch five parallel search agents for thorough coverage.

What is the difference between trace oracle, smart, and deep modes?▼

Oracle mode queries only Oracle memory and is fastest. Smart mode queries Oracle first and auto-escalates to deep mode if fewer than three results return. Deep mode launches five parallel agents covering files, git history, GitHub issues, other repos, and memory.

Can I trace a project in a remote GitHub repository?▼

Yes, pass --repo with a GitHub URL and the skill clones it into ghq before searching. Local paths also work with --repo, while omitting the flag searches the current repository.

Where are trace search results stored?▼

Results are written as timestamped markdown logs under ψ/memory/traces/YYYY-MM-DD/ and logged to Oracle MCP. Committed trace logs become searchable memory for future queries.

When should I use dig instead of trace?▼

Use dig for session mining, such as reviewing past Claude Code session timelines, since trace no longer handles session data. Trace is for locating projects, files, and commits across repos and memory.