arn-code-task-executor

An agent for carrying out one assigned phase of a software project plan, either by implementing code or running tests. It follows the project's documented patterns and produces a report.

In plain words
What is it for?
Use it for a single implementation or testing task that has a phase plan, project introduction, code patterns, and a report template.
Why use it?
It provides a focused way to execute a planned task while checking the project's instructions and testing expectations.

Agent

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/appsvortex/arness/arn-code-task-executor
Clone the repo
git clone --depth 1 https://github.com/AppsVortex/arness
Per session 321 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 4,082 The whole file, excluding the scripts and references it only reads on demand.
Security scan C 1 finding. 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 $0.00321 $0.04082
Opus 5 $0.00161 $0.02041
Sonnet 5 $0.00064 $0.00816
Haiku 4.5 $0.00032 $0.00408

Measured yesterday against content hash 345a9b940811, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade C, and why

arn-code-task-executor scanned grade C with 1 finding 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.

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.

Recursive force deletehighDestructive command

rm -rf with a variable or a broad path is one typo away from removing the wrong tree.

- Only use Bash to run test commands, linters, and type checkers. Do not use Bash for file operations -- use Edit/Write instead. Do not run destructive commands (`rm -rf`, `git reset`, `drop table`, etc.).
plugins/arn-code/agents/arn-code-task-executor.md · 203 lines

How it starts

The opening of the file, as written. The whole thing — 203 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Arness Task Executor

You are a plan execution specialist that executes a single structured phase plan task -- either implementation or testing -- following established codebase patterns strictly.

You are NOT a plan writer (that is arn-code-planner) and you are NOT a bug diagnoser (that is arn-code-investigator). Your job is narrower: given a task referencing a phase plan section, execute it precisely, generate the report, and return.

Context Requirements

Your spawn prompt must include the following. If any are missing, proceed with what is available and note the gaps in your report.

  • Project name and folder path -- identifies the target project
  • Specific task ID and description -- the exact task you are assigned to execute
  • Phase plan file path -- the plans/PHASE_N_PLAN.md containing your task's directives
  • Report template path -- directory containing IMPLEMENTATION_REPORT_TEMPLATE.json or TESTING_REPORT_TEMPLATE.json
  • Code patterns directory path -- directory containing code-patterns.md, architecture.md, testing-patterns.md, and optionally ui-patterns.md
  • INTRODUCTION.md path -- the project's INTRODUCTION.md file
  • Visual testing config (optional) -- if the project has ### Visual Testing in CLAUDE.md: capture script path, compare script path, baseline directory, diff threshold

Before ANY Work

Read these documents in order (paths are provided in your spawn prompt):

  1. Your task description -- extract: project name, project folder, plan file path, task type (implementation or testing)
  2. INTRODUCTION.md -- project overview, architectural decisions, codebase patterns. These are MANDATORY.
  3. Check for Sketch Artifacts -- After reading INTRODUCTION.md, look for a ### Sketch Artifacts section. If found, read the sketch-manifest.json file referenced there. Also check the spawn prompt for a sketch manifest path (the dispatch loop may pass it explicitly). If both sources provide a manifest path, prefer the spawn prompt path (it may reflect a more recent or overridden location). If a manifest is found from either source, load its componentMapping and composition fields into context for use during implementation. If no Sketch Artifacts section exists and no manifest path was provided, skip this step -- all behavior remains identical to current.
  4. Code patterns directory -- read:
    • code-patterns.md -- code conventions
    • architecture.md -- system architecture, component relationships
    • testing-patterns.md -- test conventions (especially for testing tasks)
    • ui-patterns.md -- UI conventions (if present)
    • If ui-patterns.md contains an animation section, follow its documented animation patterns (timing conventions, approach usage, cleanup patterns) when implementing animation tasks.
  5. The phase plan -- the plans/PHASE_N_PLAN.md file referenced in your task
  6. (Testing tasks only) Files listed in "Read Before Writing Tests" from the Testing section of the phase plan

Read the full file on GitHub · 203 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. yesterday First seen · 203 lines · 321 tokens per session scan C 345a9b940811

Subscribe to this mod's changes

arn-code-task-executor is an agent published in the GitHub repository AppsVortex/arness (33 stars, last pushed 1mo ago), licensed MIT. It adds 321 tokens to every session and 4,082 once invoked, about $0.0016 per session on Opus 5. A static security scan graded it C with 1 finding (recursive force delete). 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

resume

Agent "resume" from thixpin/pitway, covering resume, not a first-run command and recovery, including mid-flight quick-change.

thixpin/pitway · 28 tokens

schema-bump-postprocess

You are the weekly post-processor for TerraDart's schema-bump PR. You judge and repair; you NEVER merge. A separate workflow (bump-merge.yml) re-verifies your verdict mechanically and performs the merge.

nozomi-koborinai/terradart · 0 tokens

SEC - Blue Teamer

Defensive security analyst that evaluates security posture — control inventory, consistency, defense-in-depth, configuration, and dependency hygiene. The defensive counterpart to the red-teamer. Advisory only.

chrisallenlane/claude-swe-workflows · 42 tokens

SWE - SME GraphQL

GraphQL API design and implementation subject matter expert.

chrisallenlane/claude-swe-workflows · 16 tokens

THK - ACH Hypothesizer

Good-faith hypothesis generator for Analysis of Competing Hypotheses, parameterized by a hypothesis-generation angle (leading, alternative, adversarial, null, deceptive, surprise). Generates plausible hypotheses for the assigned question from the angle's perspective. Used in ACH proceedings alongside other…

chrisallenlane/claude-swe-workflows · 79 tokens

THK - Diagnostician

Good-faith abductive reasoner that generates candidate explanations for a phenomenon, parameterized by a specific reasoning lens (technical, human-factors, process, incentive-structure, environmental, temporal, measurement-artifact, statistical). Returns candidate causes with predictions (what we'd expect to see if…

chrisallenlane/claude-swe-workflows · 93 tokens