execute-project_skill

This skill loads and manages execution of existing projects, tracking progress and validating tasks to ensure steady completion.
  • Python

2

GitHub Stars

1

Bundled Files

2 months ago

Catalog Refreshed

4 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill abdullahbeam/nexus-design-abdullah --skill execute-project

  • SKILL.md20.3 KB

Overview

This skill is the only approved way to interact with existing projects. It guides project execution step-by-step, tracks progress, and validates task completion while preserving project integrity. It never reads project files directly and instead uses loader and bulk-complete scripts to inspect and update project state.

How this skill works

The skill loads project context via a loader script that returns metadata and recommended read paths, then identifies the current section and next uncompleted task by parsing task metadata rather than opening files directly. It executes work section-by-section (or task-by-task for small projects), uses bulk-complete scripts to mark tasks complete, and re-reads validated outputs through the loader to confirm progress. It supports pause/resume, incremental checkpoints, and auto-triggers a close-session when work is finished.

When to use it

  • When you say continue, resume, work on, or execute project [name/ID/number].
  • When a project status is IN_PROGRESS or PLANNING and you want to advance execution.
  • When you need systematic, validated bulk updates to task checkboxes.
  • When you want adaptive granularity for small, medium, or large projects.
  • When you need to pause and reliably resume at the next uncompleted task.

Best practices

  • Ensure planning files (overview.md, plan.md/design.md, steps.md/tasks.md) exist and tasks use checkbox format before executing.
  • Run the loader first to get recommended read paths and a project summary; avoid direct file editing during execution.
  • Follow section-based checkpoints and confirm bulk-complete actions to keep progress auditable.
  • Use adaptive granularity: task-by-task for ≤15 tasks, section checkpoints for medium projects, periodic checkpoints for large projects.
  • When pausing, optionally bulk-complete recently finished tasks so resume starts at a clean boundary.

Example use cases

  • Resume an in-progress software project to continue Implementation Section from the next uncompleted task.
  • Execute a testing phase and bulk-complete all test-case tasks after validation.
  • Pause at midday, bulk-mark recently finished tasks, trigger close-session, and resume later with preserved context.
  • Run a final bulk-complete to validate and mark a project COMPLETE, then trigger archiving workflows.
  • Use adaptive checkpoints on a large project to keep work manageable and minimize rollback risk.

FAQ

No. It uses a loader script to extract metadata and recommended reads, and uses bulk-complete scripts to update tasks without directly opening raw project files.

How does it validate progress after bulk-completing tasks?

After running bulk-complete, the skill re-queries the loader output and recommended read paths to confirm the task file shows the expected completed counts and percentages.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational
execute-project skill by abdullahbeam/nexus-design-abdullah | VeilStrat