Task Implementation with Verification

VerifiedCaution

Implements GitHub issues or standalone tasks with automatic build and test verification. Supports full mode (review + plan), plan-only, or quick execution, with auto-detection of language (Java/JavaScript) and memory-based handoff handling. Helps turn a task description or issue number into working, committed code.

Sby Skills Guide Bot
DevelopmentIntermediate
3/4/2026
Claude Code
#task-orchestration#github-issues#code-implementation#build-verification#git-workflow

Recommended for

Automated summary

This skill orchestrates implementation of GitHub issues or plain-text tasks by automatically choosing a mode (full, plan-only, or quick), then verifying the build before committing.

Strengths

  • Automatic mode selection based on whether the task is a GitHub issue number, a description, or a quick direct execution
  • Build verification loop with language detection (Maven/Java, npm/JavaScript) and up to three retries
  • State persistence through memory so an interrupted workflow can be resumed or a JSON handoff consumed
  • Built-in git commit with optional push once verification passes

Limitations

  • Depends on external skills (manage-memories, workflow-integration-git, issue-view) and the .plan/execute-script.py script being present
  • Only Java and JavaScript are supported for language detection and build verification
  • Output quality hinges heavily on how well the source issue states its acceptance criteria
When to use it

Use it when a GitHub issue or development task should be implemented end to end with build verification and a commit.

When not to use it

Avoid it for code review alone, exploratory refactoring without a verifiable build, or projects whose stack is neither Java nor JavaScript.

Security analysis

Caution
Quality score82/100

The skill is a legitimate task-implementation orchestrator, but it has powerful execution capabilities: Bash, file modification, git commit, and optional auto-push. There are no signs of destructive commands, secret exfiltration, obfuscation, or safety-disabling instructions, so it is not risky, but the broad tool access and auto-push behavior warrant caution.

Findings
  • •Uses Bash and can run arbitrary build, git, and local Python commands
  • •Can automatically commit and push changes when push=true
  • •Executes repository-local .plan/execute-script.py and delegates to other skills whose behavior is not shown
  • •Can write/edit files across the project as part of implementation
How this analysis is made

Examples

GitHub issue full workflow
/task-implement task=123
Quick edit without planning
/task-implement task="Add validation to User.java" quick
Feature description with auto-push
/task-implement task="Implement user authentication service" language=java push

name: task-implement description: Implement GitHub issues or standalone tasks with full verification user-invocable: true allowed-tools: Skill, Read, Write, Edit, Glob, Grep, Bash, Task, AskUserQuestion

Task Implementation Skill

Implement tasks through goal-based workflow with automatic mode selection and verification.

Parameters

| Parameter | Type | Description | |-----------|------|-------------| | task | required | GitHub issue number/URL or task description | | language | optional | java|javascript (auto-detects if not specified) | | quick | optional | Skip review/plan, execute directly (default: false) | | push | optional | Auto-push after successful implementation (default: false) | | handoff | optional | Handoff structure from previous task (JSON) |

Prerequisites

Load required skills:

Skill: plan-marshall:manage-memories

Workflow

Step 0: Process Handoff Input

If handoff parameter provided: Parse JSON, extract artifacts/decisions/constraints, load memory refs.

Step 1: Determine Mode

If task matches /^\d+$/ or "github.com/*/issues/" → FULL mode (Review → Plan → Execute)
If quick=true → QUICK mode (Execute only)
Otherwise → PLAN mode (Plan → Execute)

Step 2: Check Memory for Pending Workflow

python3 .plan/execute-script.py plan-marshall:manage-memories:manage-memory list --category handoffs

If pending found: Prompt "[R]esume / [S]tart fresh / [A]bort"

Step 3: Execute Mode-Specific Workflow

FULL: Load issue (via tools-integration-ci:issue-view), Review requirements, Plan implementation, Execute tasks, save progress to memory.

PLAN: Plan implementation, Execute tasks, save progress to memory.

QUICK: Execute task directly.

Step 4: Verify Implementation

Auto-detect language: pom.xml → Java, package.json → JavaScript

Run build verification. Iterate up to 3 times if fails.

Step 5: Commit and Push

If verification succeeds: Commit via git workflow.

If push=true: Run git push.

Step 6: Cleanup and Return

Cleanup memory:

python3 .plan/execute-script.py plan-marshall:manage-memories:manage-memory cleanup --category handoffs --pattern "workflow-*"

Mode Details

FULL Mode (GitHub Issue)

  1. Load issue details from GitHub
  2. Review: Analyze requirements, identify acceptance criteria
  3. Plan: Break down into implementation steps
  4. Execute: Implement each step
  5. Verify: Run build/tests

PLAN Mode (Task Description)

  1. Plan: Analyze task, create implementation steps
  2. Execute: Implement each step
  3. Verify: Run build/tests

QUICK Mode

  1. Execute: Implement task directly without planning
  2. Verify: Run build/tests

Usage Examples

Full workflow with GitHub issue:

/task-implement task=123

Quick execution (no planning):

/task-implement task="Add validation to User.java" quick

Java task with auto-push:

/task-implement task=456 language=java push

Task description with planning:

/task-implement task="Implement user authentication service"

Architecture

Delegates to skills:

/task-implement (orchestrator)
  ├─> manage-memories skill (state persistence)
  └─> workflow-integration-git skill (commit workflow)

Continuous Improvement

If you discover issues or improvements during execution, record them:

  1. Activate skill: Skill: plan-marshall:manage-lessons
  2. Record lesson with component: {type: "skill", name: "task-implement", bundle: "pm-workflow"}

Related

| Skill | Purpose | |-------|---------| | plan-marshall:manage-memories | State persistence for recovery | | pm-workflow:workflow-integration-git | Git commit workflow | | pm-workflow:pr-doctor | Fix PR issues after implementation |

From the file to a first result

The download contains exactly the content displayed on this page. This copy may differ from the latest version in the source repository.

View the source and supporting files

This download contains SKILL.md only. Any scripts, references, templates and dependencies are not included. Get the required files from the source, preserving their directory structure, before continuing.

1. Download the file

Download SKILL.md

From the root of a test project, save the file as .skill-downloads/task-implementation-with-verification-mmbizjmx/SKILL.md, or use the command below. This folder is for inspecting the skill before installation.

Command for macOS/Linux or WSL: requires a POSIX shell and curl. On Windows without WSL, use the download button and create the folder manually.

mkdir -p '.skill-downloads/task-implementation-with-verification-mmbizjmx' && curl --fail --show-error --location 'https://skills-guides.com/api/skills/task-implementation-with-verification-mmbizjmx/download' --output '.skill-downloads/task-implementation-with-verification-mmbizjmx/SKILL.md.part' && mv '.skill-downloads/task-implementation-with-verification-mmbizjmx/SKILL.md.part' '.skill-downloads/task-implementation-with-verification-mmbizjmx/SKILL.md'

2. Check prerequisites

Read SKILL.md before using it. If it references scripts or external services, inspect them and prepare the required access in your test environment.

Tools declared by the author (check availability in your agent): Skill, Read, Write, Edit, Glob, Grep, Bash, Task, AskUserQuestion

Original excerpt: Prerequisites

Load required skills:

Skill: plan-marshall:manage-memories

The platforms on this page are catalog indications. The paths below follow tool documentation; they do not prove this skill has been run in those tools. No minimum tool version is confirmed for this listing.

3. Load the skill

After inspection, copy the complete folder into your project using one of the commands below, then open a new tool session. The command stops if a folder with the same name already exists: compare the files before replacing it.

Claude Code
test -f '.skill-downloads/task-implementation-with-verification-mmbizjmx/SKILL.md' && test -s '.skill-downloads/task-implementation-with-verification-mmbizjmx/SKILL.md' && mkdir -p '.claude/skills' && test ! -e '.claude/skills/task-implement' && cp -R '.skill-downloads/task-implementation-with-verification-mmbizjmx' '.claude/skills/task-implement'

OpenSkills

OpenSkills must already be installed. The local install/read procedure is checked with OpenSkills 1.5.0, which requires Node.js ≥ 20.6.0. Catalog slugs are not resolved automatically.

Official OpenSkills installation
openskills install './.skill-downloads/task-implementation-with-verification-mmbizjmx'

openskills read 'task-implementation-with-verification-mmbizjmx'

Reading should display the installed file content and its base directory. It does not run the skill or install its dependencies.

For an agent that reads AGENTS.md, then sync the index. This command changes AGENTS.md: review the diff before using it.

openskills sync

Manual method or another tool: attach SKILL.md to a conversation, or open it in an agent that can read files. Include any required supporting resources. No universal folder is recognized by every tool.

4. Try a first task

Use a short, synthetic input: an API endpoint, a small document or an interface brief. In a new conversation, attach the file or give the agent its path, then adapt this request:

Read .skill-downloads/task-implementation-with-verification-mmbizjmx/SKILL.md and apply its instructions to this example: [describe a test input]. Identify any missing files, dependencies or access. Produce the result requested by the skill without inventing data or running unreviewed scripts.

Expected result: output matching the skill instructions, or a precise list of missing prerequisites. A read confirmation alone does not validate a first use. If an external resource is unavailable, that path remains untested.

Related skills