competition-queue-worker-drift

Trace queue payloads through async workers to prove worker-only behavior drift.

7|4|Updated Jun 22, 2026
One-click install
npx skills add https://github.com/dbx0/skills --skill competition-queue-worker-drift-dbx0
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: competition-queue-worker-drift
Source: https://github.com/dbx0/skills/tree/main/skills/web-appsec/initial-access/competition-queue-worker-drift
Command: npx skills add https://github.com/dbx0/skills --skill competition-queue-worker-drift-dbx0

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? In CTF sandbox challenges, the decisive effect often happens after enqueue, inside an async worker whose runtime differs from the request path, making bugs invisible when only the synchronous flow is analyzed. ## Core Features & Use Cases - Async Chain Mapping: Records enqueue points, queue names, payload shapes, worker consumers, retry policies, and dead-letter flows in one chain. - Worker-Only State Analysis: Compares request-time state against worker-time state including env vars, mounts, credentials, feature flags, and clocks. - Minimal Reproduction: Reduces findings to the smallest enqueue-to-side-effect sequence proving the decisive drift. - Use Case: A CTF challenge where a queued task triggers a file write or privilege-bearing action only under worker config; this skill traces the payload from enqueue through retries to the final side effect. ## Quick Start Ask the agent to trace a queue payload from the enqueue endpoint through the worker to its final side effect after the ctf-sandbox-orchestrator has established sandbox assumptions.

Frequently Asked Questions about competition-queue-worker-drift

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

FAQPage Schema
How do I trace a queue payload to its side effect in a CTF challenge?

Map the full async chain: enqueue endpoint, queue name, payload schema, worker consumer, retry policy, and final effect. Then reproduce the smallest enqueue-to-side-effect flow that proves the behavior.

Why does a task behave differently in the worker than in the request path?

Workers often run with different env vars, mounts, credentials, feature flags, or clocks than the web process. Comparing request-time state against worker-time state isolates the config drift causing the divergence.

When should I use this skill instead of analyzing the request handler?

Use it when the decisive effect happens after enqueue, inside a worker, or only under async runtime state. It is a downstream specialization that runs after the ctf-sandbox-orchestrator has established sandbox assumptions.

What evidence should I preserve when analyzing async job execution?

Keep queue names, payloads, task IDs, worker process identity, retry metadata, dedupe or lease state, and the resulting artifact or downstream mutation with timestamps. Tie enqueue side, worker side, and effect side together.

What are common pitfalls when debugging retry and dead-letter behavior?

Common mistakes include explaining only the request path, ignoring worker-only env or mount differences, and treating eventual side effects as synchronous. Always isolate the async boundary and record backoff, dedupe keys, and failure handling.