Claude Code’s ‘Duties’ replace lets brokers work longer and coordinate throughout periods

One in every of the largest constraints presently going through AI builders who need to deploy brokers in service of their particular person or enterprise objectives is the “working reminiscence” required to handle complicated, multi-stage engineering initiatives.

Sometimes, when a AI agent operates purely on a stream of textual content or voice-based dialog, it lacks the structural permanence to deal with dependencies. It is aware of what to do, however it usually forgets why it is doing it, or in what order.

With the launch of Tasks for Claude Code (introduced in v2.1.16) last week, Anthropic has launched an answer that is much less about “AI magic” and extra about sound software program engineering rules.

By shifting from ephemeral “To-dos” to persistent “Duties,” the firm is essentially re-architecting how the mannequin interacts with time, complexity, and system sources.

This replace transforms the software from a reactive coding assistant right into a state-aware venture supervisor, creating the infrastructure crucial to execute the subtle workflows outlined in Anthropic’s just-released Greatest Practices information, whereas latest changelog updates (v2.1.19) sign a spotlight on the stability required for enterprise adoption.

The structure of company: from ephemeral to persistent

To know the significance of this launch for engineering groups, we should have a look at the mechanical variations between the previous “To-do” system and the new “Process” primitive.

Beforehand, Claude Code utilized a “To-do” listing—a light-weight, chat-resident guidelines.

As Anthropic engineer Thariq Shihipar wrote in an article on X: “Todos (orange) = ‘assist Claude bear in mind what to do’.” These have been efficient for single-session scripts however fragile for precise engineering. If the session ended, the terminal crashed, or the context window drifted, the plan evaporated.

Duties (Inexperienced) introduce a brand new layer of abstraction designed for “coordinating work throughout periods, subagents, and context home windows.” This is achieved by way of three key architectural selections:

1. Dependency Graphs vs. Linear Lists: Not like a flat Todo listing, Duties help directed acyclic graphs (DAGs). A activity can explicitly “block” one other. As seen in neighborhood demonstrations, the system can decide that Process 3 (Run Exams) can’t begin till Process 1 (Construct API) and Process 2 (Configure Auth) are full. This enforcement prevents the “hallucinated completion” errors widespread in LLM workflows, the place a mannequin makes an attempt to take a look at code it hasn’t written but.

2. Filesystem Persistence & Sturdiness: Anthropic selected a “UNIX-philosophy” strategy to state administration. Moderately than locking venture state inside a proprietary cloud database, Claude Code writes duties straight to the consumer’s native filesystem (~/.claude/duties). This creates sturdy state. A developer can shut down their terminal, change machines, or recuperate from a system crash, and the agent reloads the actual state of the venture. For enterprise groups, this persistence is crucial—it means the “plan” is now an artifact that may be audited, backed up, or version-controlled, unbiased of the energetic session.

3. Orchestration through Atmosphere Variables: Probably the most potent technical unlock is the skill to share state throughout periods. By setting the CLAUDE_CODE_TASK_LIST_ID atmosphere variable, builders can level a number of situations of Claude at the identical activity listing. This permits updates to be “broadcast” to all energetic periods, enabling a degree of coordination that was beforehand unimaginable with out external orchestration instruments.

Enabling the ‘swarm’: parallelism and subagents

The discharge of Duties makes the “Parallel Classes” described in Anthropic’s Greatest Practices information sensible. The documentation suggests a Author/Reviewer sample that leverages this shared state:

1. Session A (Author) picks up Process #1 (“Implement Fee Limiter”).

2. Session A marks it full.

3. Session B (Reviewer), observing the shared state replace, sees Process #2 (“Assessment Fee Limiter”) is now unblocked.

4. Session B begins the overview in a clear context, unbiased by the era course of.

This aligns with the information’s recommendation to “fan out” work throughout recordsdata, utilizing scripts to loop by way of duties and name Claude in parallel. Crucially, patch v2.1.17 fastened “out-of-memory crashes when resuming periods with heavy subagent utilization,” indicating that Anthropic is actively optimizing the runtime for these high-load, multi-agent situations.

Enterprise readiness: stability, CI/CD, and management

For decision-makers evaluating Claude Code for manufacturing pipelines, the latest changelogs (v2.1.16–v2.1.19) reveal a spotlight on reliability and integration.

The Greatest Practices information explicitly endorses operating Claude in Headless Mode (claude -p). This permits engineering groups to combine the agent into CI/CD pipelines, pre-commit hooks, or knowledge processing scripts.

For instance, a nightly cron job may instantiate a Claude session to “Analyze the day’s log recordsdata for anomalies,” utilizing a Process listing to monitor progress by way of totally different log shards.

The transfer to autonomous brokers introduces new failure modes, which latest patches have addressed:

• Dangling Processes: v2.1.19 fastened a difficulty the place Claude Code processes would cling when the terminal closed; the system now catches EIO errors and ensures a clear exit (utilizing SIGKILL as a fallback).

• {Hardware} Compatibility: Fixes for crashes on processors with out AVX help guarantee broader deployment compatibility.

• Git Worktrees: Fixes for resume performance when working throughout totally different directories or git worktrees be sure that the “state” follows the code, not simply the shell session.

Recognizing that enterprise workflows can’t flip on a dime, Anthropic launched the CLAUDE_CODE_ENABLE_TASKS atmosphere variable (v2.1.19). Setting this to false permits groups to opt-out of the new system quickly, preserving current workflows whereas they migrate to the Process-based structure.

The builder’s workflow: managing the context financial system

For the particular person developer, the Process system solves the “context financial system” drawback. Anthropic’s documentation warns that “Claude’s context window… is the most vital useful resource to handle,” and that efficiency degrades because it fills.

Earlier than Duties, clearing the context was harmful—you wiped the agent’s reminiscence of the general plan. Now, as a result of the plan is saved on disk, customers can observe the finest observe of “aggressive context administration.” Builders can run /clear or /compact to unlock tokens for the mannequin’s reasoning, with out dropping the venture roadmap.

The changelog additionally highlights quality-of-life enhancements for energy customers constructing complicated scripts:

• Shorthand Arguments: Customers can now entry customized command arguments through $0, $1, and so on., making it simpler to script reusable “Abilities” (e.g., a /refactor command that takes a filename as an argument).

• Keybindings: Totally customizable keyboard shortcuts (/keybindings) permit for quicker interplay loops.

What Duties means for Claude Code customers

With the introduction of Duties, Anthropic is signaling that the way forward for coding brokers is a venture administration.

By giving Claude Code a persistent reminiscence, a approach to perceive dependency, and the stability fixes required for long-running processes, they’ve moved the software from a “copilot” that sits subsequent to you to a “subagent” that may be trusted to run in the background — particularly when powered by Anthropic’s most performant mannequin, Claude Opus 4.5.

It is a technical evolution that acknowledges a easy reality: in the enterprise, the code is low cost; it is the context, the plan, and the reliability that are treasured.