work-task

A command that focuses an AI assistant on one specific development task and its required deliverables. It can also load project context and check dependencies or verification requirements.

In plain words
What is it for?
Use it to work on a numbered task, read its task and issue files, include broader project context, check dependencies, and verify acceptance criteria.
Why use it?
It reduces distraction from unrelated work and provides a consistent way to track progress on an individual issue or phase.

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/agentsea/flashbacker/work-task
Clone the repo
git clone --depth 1 https://github.com/agentsea/flashbacker

Made for: Claude Code.

Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,574 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 $0.00000 $0.01574
Opus 5 $0.00000 $0.00787
Sonnet 5 $0.00000 $0.00315
Haiku 4.5 $0.00000 $0.00157

Measured 2d ago against content hash 5bec6acd8ff7, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

work-task 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 2d 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.

templates/.claude/commands/fb/work-task.md · 166 lines

How it starts

The opening of the file, as written. The whole thing — 166 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Work on Specific Atomic Task

Focus AI attention on a single atomic task with full context and clear deliverables.

Directory Structure

docs/tasks/                     # Task files location (read from here)
├── ISSUE-XXX-name/             # Per-issue task directories
│   ├── README.md               # Task progress dashboard
│   ├── phase-1-name.md         # Task file containing specific tasks
│   ├── phase-2-name.md         # Additional phase files
│   └── phase-N-name.md         # Phase files to search for task

docs/issues/                    # Issue files location (reference for context)
├── ISSUE-XXX-title.md          # Parent issue for background context
└── README.md                   # Issue tracking for dependencies

Usage

/fb:work-task <task-id> [context] [verify] [dependency-check]

Examples:

  • /fb:work-task ISSUE-027.1.1.1 - Work on template signing research task
  • /fb:work-task ISSUE-027.2.3.2 context - Include full project context for complex task
  • /fb:work-task ISSUE-015.1.2 verify - Complete task and verify acceptance criteria
  • /fb:work-task ISSUE-027.1.1.4 dependency-check - Check dependencies before starting

Command Implementation

Parse the command arguments:

  • Argument 1 (required): Task ID in format ISSUE-XXX.PHASE.CATEGORY.TASK
  • Argument 2 (optional): "context" to include full project memory and working plan
  • Argument 3 (optional): "verify" to check acceptance criteria after completion
  • Argument 4 (optional): "dependency-check" to validate prerequisites

Argument Parsing Logic

  1. Extract Task ID: First argument must match ISSUE-XXX.N.N.N pattern
  2. Parse Components: Split task ID into issue, phase, category, task numbers
  3. Check Flags: Look for "context", "verify", "dependency-check" in remaining arguments
  4. Validate Task: Ensure task exists in appropriate phase file

How It Works

Step 1: Task Loading

  1. Parse Task ID: Extract issue number, phase, category, and task number
  2. Load Task File: Read docs/tasks/ISSUE-XXX/phase-N-name.md
  3. Extract Task Details: Get specific task, deliverables, acceptance criteria
  4. Check Dependencies: Validate prerequisite tasks are marked complete

Read the full file on GitHub · 166 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. 2d ago First seen · 166 lines · 0 tokens per session scan A 5bec6acd8ff7

Subscribe to this mod's changes

work-task is a command published in the GitHub repository agentsea/flashbacker (57 stars, last pushed 7mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,574 tokens. 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.