develop_skill

This skill executes code changes from the solution package, validates KB updates, and migrates history while ensuring consistency and safety.
  • Python

352

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 hellowind777/helloagents --skill develop

  • SKILL.md12.3 KB

Overview

This skill governs the development implementation phase and enforces strict rules for executing a prepared solution package. It guides selection or confirmation of the package, performs stepwise code changes, runs security and quality checks, synchronizes the knowledge base, and migrates the executed package to history. The skill is designed for both interactive and full-authorization modes with clear failure and audit handling.

How this skill works

On entry it verifies legal preconditions (explicit confirmation, MODE_FULL_AUTH, or MODE_EXECUTION). It locates and validates a complete solution package in plan/, reads task.md and why.md, then executes tasks one-by-one with immediate status updates. The flow includes knowledge-base checks, security scans, testing with blocking/warning rules, documentation synchronization, consistency audits, and an atomic migration of the package to history/ when finished.

When to use it

  • When a solution package has been produced and awaits execution in plan/
  • When ready to apply planned code changes with automated checks and migration
  • For regulated flows that require explicit authorization or interactive confirmation
  • When you need consistent knowledge base synchronization after code changes
  • When you must enforce traceable migration of executed plans into history/

Best practices

  • Create a Git backup branch or manual backup of the code directory before execution
  • Ensure the solution package is complete (task.md, why.md, how files) before starting
  • Run in interactive mode for risky or ambiguous changes; use full-authorization for unattended runs
  • Respect single-function/class edit rule for large files and use grep/read offsets for precise edits
  • Complete knowledge base synchronization before migrating the plan to history/

Example use cases

  • Apply a multi-step feature implementation defined in plan/YYYYMMDDHHMM_feature/
  • Execute automated security and blocking tests during a release implementation
  • Run consistency audit to align documentation with code after significant refactors
  • Migrate executed plans into history/ with notes on skipped or failed tasks
  • Collect and summarize task results, failures, and changelog updates after implementation

FAQ

The skill outputs a structured error per the execution format and stops; missing files must be added before retrying.

How are blocking test failures handled?

Blocking failures halt execution immediately and present options to fix and retry, skip at risk, or terminate; auto-skip is prohibited.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational