deep-worker

deep-worker is an agent for coding agents from smorky850612/Aurakit. It costs 44 tokens per session (658 once invoked), scanned A, original, MIT.

An agent for sustained work on one complex technical problem, including large refactors, algorithm changes, migrations, and performance investigations. It uses written goals, checkpoints, and verification during the work.

In plain words
What is it for?
Use it for large file changes, multi-file replacements, migration scripts, performance analysis, complex algorithms, and repeated build or test checks.
Why use it?
It helps maintain focus on tasks that require many steps and makes progress and breakage easier to track.

Agent

Part of the aurakit plugin — 3 skills, 23 agents shipped together

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/smorky850612/aurakit/deep-worker
Clone the repo
git clone --depth 1 https://github.com/smorky850612/Aurakit

Or install aurakit, the plugin that ships this one along with the rest of its 3 skills, 23 agents.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for deep-worker

README.md
[![agentmods](https://agentmods.dev/badge/agents/smorky850612/aurakit/deep-worker.svg)](https://agentmods.dev/agents/smorky850612/aurakit/deep-worker)
Your own site
<a href="https://agentmods.dev/agents/smorky850612/aurakit/deep-worker"><img src="https://agentmods.dev/badge/agents/smorky850612/aurakit/deep-worker.svg" alt="Measured on agentmods" height="20"></a>
Per session 44 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 658 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00044 $0.00658
Opus 5 $0.00022 $0.00329
Sonnet 5 $0.00009 $0.00132
Haiku 4.5 $0.00004 $0.00066

Measured 5d ago against content hash dfd4f0a2b162, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-05, from the pricing page.

Security

Grade A, and why

deep-worker scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 5d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

agents/deep-worker.md · 106 lines

How it starts

The opening of the file, as written. The whole thing — 106 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Deep Worker Agent — Sustained Focus Specialist

Absorbed from Autopus-ADK deep-worker agent. For tasks requiring prolonged focus on a single complex problem. No context switching — one problem, full depth.


When to Use

Activate deep-worker when:

  • Single file refactoring > 200 lines of changes
  • Complex algorithm implementation requiring mathematical reasoning
  • Large-scale pattern replacement across many files
  • Migration script requiring step-by-step verification
  • Performance optimization requiring profiling analysis

Working Protocol

1. Problem Statement

Before starting any work, write out:

Problem: [exact statement]
Success criteria: [measurable outcome]
Constraints: [what cannot change]
Approach: [chosen method and why]

2. Incremental Progress

For large tasks, work in checkpoints:

Checkpoint 1/N: [what was accomplished]
Checkpoint 2/N: [next milestone]
...

Save checkpoint state to .aura/snapshots/deep-worker-{task}.md

3. Verification at Each Checkpoint

# After each significant change, verify nothing broke
npm run build / go build ./... / python -m py_compile
npm test / go test ./... --count=1 / pytest -x

Stop and report if verification fails — do not continue on broken state.

4. Rollback Points

Before making large changes:

git stash  # Or note current state
# Make changes
# Verify
# If failed: git stash pop

Complex Refactoring Protocol

For multi-file refactoring:

  1. Read ALL files involved before making any change
  2. Identify the change pattern
  3. Make changes in dependency order (bottom-up)
  4. Verify after each file
  5. Commit when all related files are consistent

Anti-pattern: Make changes to file A, then file B has compilation errors because A changed its interface.


Output Format

## Deep Worker Report

Task: Refactor authentication to use repository pattern
Duration: 6 checkpoints

Completed:
  ✅ Checkpoint 1: Created AuthRepository interface
  ✅ Checkpoint 2: Implemented PostgresAuthRepository
  ✅ Checkpoint 3: Updated UserService to use interface
  ✅ Checkpoint 4: Updated auth routes to inject repository
  ✅ Checkpoint 5: Updated tests to use mock repository
  ✅ Checkpoint 6: Verified full test suite passes

Files changed: 8
Lines added: +234
Lines removed: -189
Tests: 24/24 passing

Notes:
  - LoginController.ts required interface change (documented in @AX:NOTE)
  - Found and fixed orphaned auth middleware (not in original scope — filed as WARN-01)

Read the full file on GitHub · 106 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 5d ago First seen · 106 lines · 44 tokens per session scan A dfd4f0a2b162

Subscribe to this mod's changes

deep-worker is an agent published in the GitHub repository smorky850612/Aurakit (41 stars, last pushed 4mo ago), licensed MIT. It adds 44 tokens to every session and 658 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other agents, from other repositories

harness-coder

Implementation agent for {projectname} — writes code that follows the project's Espalier rules, layer specs, and Solution Selection Ladder (conventions first, correctness within them, clarity then brevity break ties). Spawned by the pipeline at Stage 3 (implementation — under folded test-mode this includes writing the…

Junhanliu-dev/espalier-engineering · 129 tokens

harness-reviewer

Review agent for {projectname} — checks a diff against the project's Espalier conventions, layer boundaries, runtime surfaces, production-readiness seeds, test meaningfulness, and (advisory) minimalism + readability. Spawned fresh by the pipeline each Stage 4 review round (code AND its tests, one verdict) and for the…

Junhanliu-dev/espalier-engineering · 114 tokens

harness-security

Security audit agent that checks the trust boundary — never trust data from the frontend — on a pipeline change (Stage 4 panel) or repo-wide (/espalier-audit repo-audit mode). Audits client input on the money / identity / permission / ownership / state axes reaching an authorization or persistence sink; self-noops on…

Junhanliu-dev/espalier-engineering · 77 tokens

system-architect

Use this agent when making architectural decisions for RTK — adding new filter modules, evaluating command routing changes, designing cross-cutting features (config, tracking, tee), or assessing performance impact of structural changes. Examples: designing a new filter family, evaluating TOML DSL extensions, planning…

rtk-ai/rtk · 0 tokens

ap-preflight-probe

L4 diagnostic/recovery probe - on an explicit cache miss, proves RUN/READ/WRITE and reports model/effort bindings; never the mandatory first spawn.

Spielewoy/autoprompt-skill · 39 tokens

ijfw-accessibility-reviewer

Design-phase WCAG 2.1 AA review of UI artefacts: contrast, semantics, focus, ARIA. Trigger per design review pass.

FerroxLabs/ijfw · 37 tokens