Trace And Isolate

What it does

When your agent uses it

• A bug is hard to locate
• Code is not working as expected
• Crash appears with unclear cause
• A regression was introduced between recent commits

Example prompts

• “Use the trace-and-isolate skill to apply systematic tracing and isolation techniques to pinpoint exactly where a bug originates in code”
• “/trace-and-isolate”

Workflow steps

4 steps, taken from the first numbered list in SKILL.md.

1. 1Identify the start - Last known good state
2. 2Identify the end - First observed bad state
3. 3Test the middle - Check if behavior is good or bad
4. 4Repeat - Binary search on the half with the bug

What it can do on your machine

Read from SKILL.md and the folder at commit d2e5c34. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  Shell commands in SKILL.md call:

  • git

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  No URLs in SKILL.md. Its commands use git, which can reach the network depending on how they are called.

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names no API keys, tokens, secrets or passwords.

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Trace And Isolate loads about 1.8k tokens when it runs. Until then it costs about 155 tokens; SKILL.md has 378 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~155

When it runs · the whole SKILL.md, loaded when a task matches

~1.8k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from rohitg00/skillkit at commit d2e5c34, republished under its Apache-2.0 licence (© rohitg00). 378 words, ~1,825 tokens.

name

trace-and-isolate

description

Applies systematic tracing and isolation techniques to pinpoint exactly where a bug originates in code. Use when a bug is hard to locate, code is not working as expected, an error or crash appears with unclear cause, a regression was introduced between recent commits, or you need to narrow down which component, function, or line is faulty. Covers binary search debugging, git bisect for regressions, strategic logging with [TRACE] patterns, data and control flow tracing, component isolation, minimal reproduction cases, conditional breakpoints, and watch expressions across TypeScript, SQL, and bash.

version

1.0.0

triggers

trace execution, isolate bug, narrow down, binary search debugging, find where

tags

debugging, tracing, isolation, bisect

difficulty

intermediate

estimatedTime

15

relatedSkills

debugging/root-cause-analysis, debugging/hypothesis-testing

Start: User clicks submit button
End: Error shown to user

Midpoint 1: Form validation
→ Data looks correct here? Continue to later code
→ Data already wrong? Focus on earlier code

Midpoint 2: API request construction
→ Request payload correct? Focus on server side
→ Payload already malformed? Focus on form handling

...Continue until exact line is found

# Start bisect
git bisect start

# Mark current (broken) state as bad
git bisect bad

# Mark last known good state
git bisect good v2.3.0

# Git checks out middle commit, test and mark
git bisect good  # or git bisect bad

# Repeat until found
# Git will report: "abc123 is the first bad commit"

# Clean up
git bisect reset

git bisect start HEAD v2.3.0
git bisect run npm test -- --grep "failing test"

function processOrder(order) {
  console.log('[TRACE] processOrder input:', JSON.stringify(order));

  const validated = validateOrder(order);
  console.log('[TRACE] after validation:', JSON.stringify(validated));

  const priced = calculatePrice(validated);
  console.log('[TRACE] after pricing:', JSON.stringify(priced));

  const result = submitOrder(priced);
  console.log('[TRACE] final result:', JSON.stringify(result));

  return result;
}

Input: { userId: "123", items: [...] }
    ↓
validateUser() → { userId: "123", verified: true }
    ↓
enrichItems() → { userId: "123", verified: true, items: [...enriched] }
    ↓
calculateTotals() → { ..., subtotal: 100, tax: 8, total: 108 }
    ↓
Output: { orderId: "456", total: 108 }

function handleRequest(req) {
  console.log('[TRACE] handleRequest entered');

  if (req.authenticated) {
    console.log('[TRACE] authenticated path');
    if (req.isAdmin) {
      console.log('[TRACE] admin path');
      return handleAdminRequest(req);
    } else {
      console.log('[TRACE] user path');
      return handleUserRequest(req);
    }
  } else {
    console.log('[TRACE] unauthenticated path');
    return handlePublicRequest(req);
  }
}

Full System: Frontend → API → Database
              ↓
Test 1: Frontend → Mock API
        → Works? Problem is in API or Database

Test 2: Real API → Mock Database
        → Works? Problem is in Database

Test 3: API with minimal data
        → Works? Problem is data-dependent

function complexFunction(input) {
  // BREAKPOINT 1: Entry - check input
  const step1 = transform(input);
  // BREAKPOINT 2: After first transformation

  for (const item of step1.items) {
    // BREAKPOINT 3: Inside loop - conditional on item
    process(item);
  }

  // BREAKPOINT 4: Exit - check output
  return finalize(step1);
}

-- Create isolated test data
BEGIN TRANSACTION;
-- Insert test data
-- Run test queries
-- Verify results
ROLLBACK;

// Intercept and log all network requests
const originalFetch = window.fetch;
window.fetch = async (...args) => {
  console.log('[TRACE] fetch:', args[0]);
  const response = await originalFetch(...args);
  console.log('[TRACE] response:', response.status);
  return response;
};

// Control time for debugging
const realNow = Date.now;
Date.now = () => {
  const time = realNow();
  console.log('[TRACE] Date.now():', new Date(time).toISOString());
  return time;
};

© rohitg00, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in packages/core/src/methodology/packs/debugging/trace-and-isolate of rohitg00/skillkit.

Open the folder on GitHubat commit d2e5c34