A skill that delegates a project task to a child coding-agent process and supervises its work. It uses task files, progress records, status data, and review records to track the run.

Use it to launch, monitor, and review a delegated coding task, including its plan, progress, evidence, and final outputs.

It gives longer or substantial tasks a repeatable workflow with visible progress and review checkpoints. It also encourages checking whether configuration or simplification can solve the problem before adding code.

task-runner — Skill by rdudov

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 skills/rdudov/task-agent/task-runner

Any agent

npx skills add rdudov/task-agent --skill task-runner

Clone the repo

git clone --depth 1 https://github.com/rdudov/task-agent

Made for: Claude Code, Codex.

Per session 73 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 15,481 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.

Model	Per session	Once invoked
Fable 5	$0.00073	$0.15481
Opus 5	$0.00036	$0.07741
Sonnet 5	$0.00015	$0.03096
Haiku 4.5	$0.00007	$0.01548

Measured yesterday against content hash 7640e3c3d3a6, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

task-runner 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 yesterday.

The scan reads SKILL.md. This mod also ships 29 executable files (scripts/__init__.py, scripts/application_adapter.py, scripts/dev_pipeline_adapter.py, …), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

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.

skills/task-runner/SKILL.md · 1,107 lines

How it starts

The opening of the file, as written. The whole thing — 1,107 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Task Runner

This skill launches a child CLI agent to execute a task from its task directory.

Use the standard single-child workflow by default. Use the dev-pipeline workflow when a task should run through the evidence-gated dev-pipeline lifecycle instead.

Before implementation, use the canonical no-code-first order: do nothing; remove or disable; configure or reuse; simplify; only then add the smallest necessary code and state why the observed gap required it. Independent review treats avoidable code or needless complexity as a defect.

Artifacts

The runner expects:

task.md
plan.md

It creates or updates:

trace.md
status.json
.runner/prompt.txt
.runner/runner.json
.runner/runner.log
.runner/review-admission.json for the current launch, .runner/review-admission-commitment.json while a committed binding is waiting for its author to start, and reviews/admissions.jsonl for every admission this number has made
reviews/rounds.jsonl and reviews/infrastructure-obligations.jsonl, once a review round or a review outage happens

The child is asked to publish progress.json for long runs, and to place explicitly requested output files in deliverables/ with deliverables/manifest.json.

The dev-pipeline workflow also creates dev-pipeline/.

An application that already supervises the entire worker service or container may use start --foreground and review --foreground. These forms keep the same admission, binding, phases, review rounds and completion decision, but wait for the child in the caller process instead of detaching a watcher that the outer container would terminate. The recorded supervision durability is caller_owned; foreground never supplies a missing assurance strategy or reviewer.

Runner Selection

The child runner follows the parent CLI agent. task_runner.py owns that decision, so callers that never read this file get the same resolution. The order is:

an explicit --runner codex|claude|agent, which always wins;
the TASK_AGENT_CHILD_RUNNER environment variable;
the nearest claude or codex process ancestor;
a single vendor's session markers in the environment (CODEX_THREAD_ID, CLAUDECODE), used only when no CLI ancestor is visible;
the documented fallback codex.

Read the full file on GitHub · 1,107 lines

Files

What ships with it

29 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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.

yesterday First seen · 1,107 lines · 73 tokens per session scan A 7640e3c3d3a6

Subscribe to this mod's changes

task-runner is a skill published in the GitHub repository rdudov/task-agent (4 stars, last pushed 2d ago), licensed MIT. It adds 73 tokens to every session and 15,481 once invoked, about $0.0004 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.

Code review Planning

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

next-cache-components-adoption

Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…

vercel/next.js · 95 tokens

babysit-pr

Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…

openai/codex · 114 tokens

imagegen

Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…

openai/codex · 113 tokens

cpu-profile-analysis

Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…

microsoft/vscode · 71 tokens

next-cache-components-optimizer

Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…

vercel/next.js · 170 tokens