marketplace/bundles/pm-plugin-development/skills/plugin-plan-implement/SKILL.md
Implement plugin tasks from plan with step iteration and progress tracking
npx skillsauth add cuioss/plan-marshall plugin-plan-implementInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Role: Implement plugin-domain tasks by iterating through steps (file paths) and applying changes.
Execution Pattern: Load task → Load skills → Iterate steps → Apply changes → Verify → Return result
Skill: plan-marshall:dev-agent-behavior-rules
| Script | Purpose |
|--------|---------|
| plan-marshall:manage-tasks:manage-tasks | Task retrieval and progress tracking |
| plan-marshall:manage-logging:manage-logging | Work log entries |
| plan-marshall:manage-config:manage-config | Domain skill retrieval |
Read standards/step-execution.md
Contains: How to execute each step type (modify, create, etc.)
| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| plan_id | string | Yes | Plan identifier |
| task_number | number | Yes | Task to execute |
python3 .plan/execute-script.py plan-marshall:manage-tasks:manage-tasks read \
--plan-id {plan_id} \
--task-number {task_number}
Extract from response:
title: Task title for loggingdescription: What changes to applydelegation.domain: For loading default skillsdelegation.context_skills: Additional skillssteps[]: File paths to processverification: Commands and criteriapython3 .plan/execute-script.py plan-marshall:manage-config:manage-config \
skill-domains get-defaults \
--domain {delegation.domain}
Skill: {default_skill_1}
Skill: {default_skill_2}
From task delegation block:
Skill: {delegation.context_skills[0]}
Skill: {delegation.context_skills[1]}
python3 .plan/execute-script.py plan-marshall:manage-logging:manage-logging \
work --plan-id {plan_id} --level INFO --message "[TASK] (pm-plugin-development:plugin-plan-implement) Starting task {task_number}: {title}"
For each step WHERE status == pending:
The step title is a file path. Apply changes based on:
description - overall change guidanceFor plugin files (agents, commands, skills):
Load step execution patterns if needed:
Read standards/step-execution.md
python3 .plan/execute-script.py plan-marshall:manage-logging:manage-logging \
work --plan-id {plan_id} --level INFO --message "[STEP] (pm-plugin-development:plugin-plan-implement) Completed step {step_number}: {file_path}"
python3 .plan/execute-script.py plan-marshall:manage-tasks:manage-tasks finalize-step \
--plan-id {plan_id} \
--task-number {task_number} \
--step {step_number} \
--outcome done
After all steps complete:
# Execute each verification command
{verification.commands[0]}
{verification.commands[1]}
Log result:
python3 .plan/execute-script.py plan-marshall:manage-logging:manage-logging \
work --plan-id {plan_id} --level INFO --message "[VERIFY] (pm-plugin-development:plugin-plan-implement) Verification {passed|failed}: {criteria}"
Success Output:
status: success
plan_id: {plan_id}
task_number: {task_number}
execution_summary:
steps_completed: {N}
steps_total: {M}
files_modified[N]:
- {path1}
- {path2}
verification:
passed: true
command: "{verification command}"
next_action: task_complete
Error Output:
status: error
plan_id: {plan_id}
task_number: {task_number}
execution_summary:
steps_completed: {N}
steps_failed: {M}
failure:
step: {step_number}
file: "{file path}"
error: "{error message}"
recoverable: true
next_action: requires_attention
If a step fails:
If verification fails:
recoverable: trueplan-marshall:execution-context-{level} - Generic execution-context dispatcher (loaded via the workflow prompt-body field)pm-plugin-development:plugin-architecture - Architecture principlesplan-marshall:manage-tasks - Task and step managementplan-marshall:manage-logging:manage-logging - Work loggingplan-marshall:phase-5-execute - Generic plan execution orchestratorpm-plugin-development:plugin-maintain - Plugin maintenance operationstesting
A test skill for README generation
testing
A test skill with existing references
tools
Skill without references directory
development
Test skill with table-format references