deepwork

deepwork is a command for Claude Code from ils15/pantheon-legacy. It costs 34 tokens per session (873 once invoked), scanned A, original, MIT.

A workflow for large coding tasks split into several phases, with saved checkpoints and reviews between phases. It can resume work after an interruption or across multiple sessions.

In plain words
What is it for?
Use it for complex changes expected to take many turns, span sessions, or require separate discovery, implementation, review, and verification stages.
Why use it?
It reduces the risk of losing progress or context during long tasks with dependent steps and multiple coding agents.

Command for Claude Code

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 commands/ils15/pantheon-legacy/deepwork
Clone the repo
git clone --depth 1 https://github.com/ils15/pantheon-legacy

Made for: Claude Code.

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 deepwork

README.md
[![agentmods](https://agentmods.dev/badge/commands/ils15/pantheon-legacy/deepwork.svg)](https://agentmods.dev/commands/ils15/pantheon-legacy/deepwork)
Your own site
<a href="https://agentmods.dev/commands/ils15/pantheon-legacy/deepwork"><img src="https://agentmods.dev/badge/commands/ils15/pantheon-legacy/deepwork.svg" alt="Measured on agentmods" height="20"></a>
Per session 34 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 873 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 $0.00034 $0.00873
Opus 5 $0.00017 $0.00436
Sonnet 5 $0.00007 $0.00175
Haiku 4.5 $0.00003 $0.00087

Measured 4d ago against content hash 408e5e11f070, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

deepwork 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 4d 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.

Origin

Copies of this mod

1 near-identical copy found in the catalogue:

.claude/commands/deepwork.md · 102 lines

How it starts

The opening of the file, as written. The whole thing — 102 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/deepwork — Heavy Task Workflow

What: Starts a structured, checkpointed workflow for complex multi-step tasks. Progress is persisted to .pantheon/deepwork/<task-slug>/ so work can resume if interrupted. Each phase is gated by Themis review.

When to Use

  • Tasks expected to take 10+ turns
  • Multi-agent orchestration with dependencies
  • Work that spans multiple sessions
  • Anything where losing context mid-way would be costly

When NOT to Use

  • Simple fixes (use @talos)
  • Single-file changes (use /subtask)
  • Tasks completable in < 5 turns (use normal delegation)

Workflow

Phase 0: SCOPING
  └─ Zeus + Athena define task scope, phases, and acceptance criteria
  └─ Output: .pantheon/deepwork/<slug>/PLAN.md

Phase 1: DISCOVERY
  └─ @apollo maps relevant codebase areas
  └─ Output: .pantheon/deepwork/<slug>/DISCOVERY.md

Phase 2-N: IMPLEMENTATION (parallel per phase)
  └─ @hermes / @aphrodite / @demeter implement phase scope
  └─ Output: .pantheon/deepwork/<slug>/phase-<N>-<agent>.md

GATE after each phase:
  └─ @themis reviews phase output
  └─ FAIL → fix and re-submit
  └─ PASS → continue to next phase

Phase FINAL: VERIFICATION
  └─ @themis full integration review
  └─ All tests pass, coverage >80%
  └─ Output: .pantheon/deepwork/<slug>/REVIEW.md

Checkpoint System

Each phase writes a checkpoint file. If work is interrupted:

/deepwork --resume <slug>    # Resume from last checkpoint
/deepwork --status <slug>    # Show current progress
/deepwork --list             # List all deepwork tasks

Checkpoint files:

.pantheon/deepwork/<slug>/
├── PLAN.md                  # Scope and phase plan
├── DISCOVERY.md             # Codebase map
├── phase-1-hermes.md        # Phase 1 backend work
├── phase-1-aphrodite.md     # Phase 1 frontend work
├── phase-1-review.md        # Phase 1 Themis review
├── phase-2-*.md             # Subsequent phases
├── REVIEW.md                # Final Themis review
└── STATUS.md                # Current state (phase, progress, blockers)

Read the full file on GitHub · 102 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. 4d ago First seen · 102 lines · 34 tokens per session scan A 408e5e11f070

Subscribe to this mod's changes

deepwork is a command published in the GitHub repository ils15/pantheon-legacy (10 stars, last pushed yesterday), licensed MIT. It adds 34 tokens to every session and 873 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 2026-08-31.