Agent skill

ctx

Use when starting a session on a project, returning after time away, or before making significant changes. Essential for building comprehensive understanding of project state through total recall and deep exploration.

Stars 3
Forks 1

Install this agent skill to your Project

npx add-skill https://github.com/Cygnusfear/claude-stuff/tree/main/skills/ctx

SKILL.md

Contextualize

You need to fully contextualize yourself in this repository before doing any work.

Step 1: Search Total Recall

First, search Total Recall for relevant context about this project. Use whichever tool name is available:

# Claude Code:
mcp__plugin_totalrecall_memory-total-recall__synthesis_search({
  query: "project name/key terms recent work decisions made",
  max_results: 15
})

# Codex/other:
totalrecall.synthesis_search({
  query: "project name/key terms recent work decisions made",
  max_results: 15
})

Look for:

  • Recent conversations about this project
  • Decisions that were made
  • Problems that were solved
  • Patterns established
  • Gotchas discovered

Step 2: Dispatch the Oracle for Deep Context

If Task tool is available, use it to dispatch an Oracle:

Task(
  subagent_type: "general-purpose",
  model: "opus",
  prompt: """
  You are The Oracle - contextualize yourself deeply in this repository.
  [see investigation checklist below]
  """
)

If Task tool is NOT available, perform the investigation yourself directly.

Investigation Checklist

Whether delegated to Oracle or done directly, cover these areas:

  1. Project Identity

    • Read CLAUDE.md, README.md, package.json (or equivalent)
    • What is this project? What does it do?
    • What are the core technologies and patterns?
    • What are the established conventions?
  2. Current Work Status

    • Check .plans/ directory for active plans
    • Check .tasks/ or TODO files for pending work
    • Check git status and recent commits (git log -10 --oneline)
    • What was being worked on? What's in progress?
  3. Decision History

    • Check .plans/DECISIONS.md or .plans/decisions/
    • What major decisions have been made?
    • What constraints exist from past choices?
  4. Architecture Overview

    • Map the key directories and their purposes
    • Identify core modules/services
    • Understand the data flow
  5. Pain Points & Context

    • Look for TODOs, FIXMEs, HACKs in code
    • Check for any KNOWN_ISSUES or TROUBLESHOOTING docs
    • Identify technical debt markers

Deliverable Format

markdown
## Project Summary
[What this is, in 2-3 sentences]

## Current Status
- Active work: [what's in progress]
- Blocked items: [if any]
- Recent changes: [last few commits summary]

## Key Decisions to Remember
- [Decision 1]
- [Decision 2]

## Established Patterns
- [Pattern 1]: [where/how used]
- [Pattern 2]: [where/how used]

## Conventions & Constraints
- [Convention from CLAUDE.md]
- [Technical constraint]

## Recommended Next Steps
Based on current state, suggest what to focus on.

Step 3: Synthesize and Report

After gathering context from memory and investigation:

  1. Summarize the key findings for the user
  2. Highlight anything that seems urgent or blocked
  3. Confirm understanding of current priorities
  4. Ask if there's specific context you're missing

When to Use This Command

  • Starting a new session on a project
  • Returning to a project after time away
  • Before making significant changes
  • When feeling uncertain about project state
  • After another developer has made changes

Remember: Context prevents mistakes. Take the time to understand before acting.

Expand your agent's capabilities with these related and highly-rated skills.

Cygnusfear/claude-stuff

brainstorming

You MUST use this, UNLESS the human says otherwise, before any EXTENDED creative work (not simple execution) - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.

3 1
Explore
Cygnusfear/claude-stuff

axiom-audit

ONLY USE WHEN REPO IS USING AXIOM LOGGING. Audit Axiom logs to identify and prioritize errors and warnings, research probable causes, and flag log smells. Use when user asks to check Axiom logs, analyze production errors, investigate log issues, or audit logging patterns.

3 1
Explore
Cygnusfear/claude-stuff

test-driven-development

Use when implementing any feature or bugfix, before writing implementation code

3 1
Explore
Cygnusfear/claude-stuff

create-plan

Create comprehensive implementation plan in .plans directory based on analysis or report. Use when user asks to create a plan, plan implementation, design a solution, or structure work for a feature/refactor/fix.

3 1
Explore
Cygnusfear/claude-stuff

video-explorer

This skill should be used when analyzing video files. Claude cannot process video directly, so this skill extracts frames hierarchically - starting with a quick overview, then zooming into regions of interest with higher resolution and temporal density. Use when asked to watch, analyze, review, or understand video content.

3 1
Explore
Cygnusfear/claude-stuff

using-superpowers

Guidelines for using skills effectively - load relevant skills before complex tasks, not every message

3 1
Explore

Didn't find tool you were looking for?

Be as detailed as possible for better results