- Home
- Skills
- Gohypergiant
- Agent Skills
- Skill Template
skill-template_skill
- TypeScript
6
GitHub Stars
3
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 gohypergiant/agent-skills --skill skill-template- AGENTS.md2.9 KB
- README.md3.2 KB
- SKILL.md6.9 KB
Overview
This skill packages expert guidance for creating and operating reproducible shell-based agent workflows. It focuses on decision trees, failure modes, and practical checks that prevent common, hard-to-debug mistakes. Use it to harden scripts, choose safe defaults, and structure progressive guidance for agents and maintainers.
How this skill works
The skill inspects workflow boundaries, fragile file operations, and places where human judgment is required. It surfaces domain-specific anti-patterns, prompts for critical pre-action questions, and provides a phased decision process to pick safe commands or fallbacks. Outputs are concise checks and exact-action recommendations tailored for shell-centric automation.
When to use it
- Designing or reviewing shell scripts that modify files, deployments, or system state
- Building agent workflows that must balance autonomy with safety
- Onboarding contributors to an existing automation pipeline
- Creating templates where low-risk defaults and explicit failure modes are required
- Auditing operations that could cause data corruption or irreversible changes
Best practices
- NEVER run destructive commands without explicit, recorded confirmation — irreversible side effects are common and often unnoticed
- NEVER assume canonical paths or environment variables — validate and normalize inputs to avoid cross-system failures
- NEVER edit compressed or packaged files in place — unpack, modify, and repack to prevent corruption
- NEVER rely solely on heuristics for backups — use deterministic versioned snapshots or checksums before changes
- Ask targeted pre-action questions to determine scope, rollback strategy, and acceptable risk before executing
- Prefer small, testable steps with clear failure signals and automatic rollbacks where possible
Example use cases
- Safely migrate configuration files across environments with pre-checks and verification steps
- Build an agent flow that runs maintenance scripts only after automated sanity checks and operator confirmation
- Create a deployment script that performs atomic file swaps and keeps previous versions for quick rollback
- Audit a CI/CD pipeline to replace fragile heuristics with deterministic checksums and explicit invariants
- Train new maintainers with a checklist-driven workflow that minimizes destructive trial-and-error
FAQ
Confirm exact targets, backup availability, rollback steps, and acceptable outage window; if any are unknown, pause and collect more information.
How do I balance automation with safety?
Automate safe, idempotent checks and low-risk steps; require explicit human approval for irreversible actions and provide automated rollback paths.