From arn-code
This skill should be used when the user says "execute task N", "run task N", "implement task N", "re-run task N", "retry task N", "run single task", or wants to execute a single specific task from the task list with optional review. This is for ONE task only — for executing the full plan (all tasks), use arn-code-execute-plan instead.
npx claudepluginhub appsvortex/arness --plugin arn-codeThis skill uses the workspace's default tool permissions.
Searches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.
Searches prompts.chat for AI prompt templates by keyword or category, retrieves by ID with variable handling, and improves prompts via AI. Use for discovering or enhancing prompts.
Compares coding agents like Claude Code and Aider on custom YAML-defined codebase tasks using git worktrees, measuring pass rate, cost, time, and consistency.
Execute a single task from the task list by spawning a arn-code-task-executor agent. Optionally run a arn-code-task-reviewer to validate the implementation before marking the task complete.
Pipeline position (this is an alternative entry point, not in the main pipeline):
arn-code-taskify -> arn-code-execute-task (single task) -> [optional] arn-code-review-implementation
Use cases:
If no ## Arness section exists in the project's CLAUDE.md, inform the user: "Arness is not configured for this project yet. Run /arn-implementing to get started — it will set everything up automatically." Do not proceed without it. Task list must exist (run /arn-code-taskify first).
Read the project's CLAUDE.md and extract the ## Arness section to find:
ask, auto, or manual (if present)Template version check: If Template version and Template updates fields are present, run the template version check procedure documented in arn-code-save-plan Step 1 (Template Version Check) before proceeding. If ## Arness does not contain these fields, treat as legacy and skip.
Ask (using AskUserQuestion):
"Run with review gate after execution?"
Options:
If review chosen, also ask:
Ask (using AskUserQuestion):
"How should Arness handle critical review findings?"
Options:
Store the choice for the execution session.
Ask the user for PROJECT_NAME if not provided in the trigger message
Verify the project directory exists:
<plans-dir>/<PROJECT_NAME>/
├── INTRODUCTION.md
├── TASKS.md
├── PROGRESS_TRACKER.json
├── plans/PHASE_*.md
└── reports/
If PROGRESS_TRACKER.json is missing, warn that progress tracking will not be available. Execution can still proceed.
Mark task as in_progress via TaskUpdate
Check if the project has ### Visual Testing configured in the ## Arness section of CLAUDE.md. If found, extract: capture script path, compare script path, baseline directory, diff threshold.
Note: Extract only the top-level fields (implicit Layer 1). Do NOT parse #### Layer N: subsections — multi-layer visual validation runs during /arn-code-review-implementation, not per-task.
Spawn arn-code-task-executor via the Task tool with full context:
<PROJECT_NAME><plans-dir>/<PROJECT_NAME>/<project-folder>/INTRODUCTION.md<code-patterns-dir>/<specs-dir>/ (if relevant to the task)<template-path>/Record the agent ID returned by the Task tool (needed for resume mode)
Wait for executor to complete
arn-code-task-reviewer via the Task tool with:
/arn-code-review-implementation.reviewFindings section), mark task as completed via TaskUpdateresume parameter of the Task tool) with review feedback, then re-run reviewerPROGRESS_TRACKER.json:
<plans-dir>/<PROJECT_NAME>/PROGRESS_TRACKER.jsonimplementation.taskId or testing.taskId matches this task"completed" and completedAt to current ISO 8601 timestampreview.verdict to the reviewer's verdict and review.reviewCycles to the number of review cyclesimplementation.status = "completed" (and testing.status = "completed" or "none"), set overallStatus to "completed"; otherwise set overallStatus to "in_progress" (if not already)lastUpdated and write to diskreview.verdict and review.reviewCycles at their initial valuesPROGRESS_TRACKER.json does not exist, skip this step silentlyShow: what was implemented, tests run, review verdict, reports generated.
Offer next step: "Would you like to run /arn-code-review-implementation to validate the full project, or continue with another task?"
If the review verdict is pass or pass-with-warnings: also offer "Or run /arn-code-simplify to review the implementation for simplification opportunities before full review."
If deferred visual testing layers exist in CLAUDE.md (any #### Layer N: subsection with **Status:** deferred) and this task is the last task in a phase, suggest: "Deferred visual testing layers detected. Consider running /arn-spark-visual-readiness to check if they can be activated now that this phase is complete."
## Arness config missing in CLAUDE.md -- suggest running /arn-implementing to get started/arn-code-taskify to convert TASKS.md into tasks/arn-code-save-plan to create the project structure