Hyvän mielen pelit ry
Official@hyvanmielenpelit · Helsinki, Finland
Hyvän mielen pelit ry publishes conventions for orchestrating coding agents across Claude Code, Codex, and Gemini Antigravity, covering planning, subagents, quotas, and secret hygiene.
Agent Skills by Hyvän mielen pelit ry
Showing 10 vetted skills indexed across 1 GitHub repositories.
hyvanmielenpelitcodex-conventions
Defines Codex-specific conventions for plan delivery, sandbox escalation, and subagent delegation.
hyvanmielenpelitclaude-plan-mode
Reconciles Claude Code plan mode with the shared plans repository workflow.
hyvanmielenpelitclaude-code-conventions
Defines Claude Code conventions for subagent spawning, skill stubs, and document linking.
hyvanmielenpelitclaude-usage-quota-budget
Measures Claude Code subscription quota consumption from local transcripts to stop plan execution at step boundaries.
hyvanmielenpelitagent-powershell-guidelines
Guides AI agents in executing PowerShell commands and file I/O safely on Windows.
hyvanmielenpelitagent-implementation-planning
Defines a five-phase workflow for researching, approving, executing, and verifying implementation plans.
hyvanmielenpelitagent-subagent-guidelines
Defines rules for assigning implementation work to AI subagents within approved plans.
hyvanmielenpelitagent-secret-hygiene
Prevents secret-shaped literals in code, tests, and documentation from triggering GitHub push protection.
hyvanmielenpelitgemini-skill-inclusion
Explains Antigravity skill discovery and skills.json configuration for global skill registration.
hyvanmielenpelitgemini-antigravity-conventions
Applies Antigravity app conventions for plan delivery, model tiers, and skill discovery.
Frequently Asked Questions About Hyvän mielen pelit ry
FAQPage SchemaWhat tasks can I accomplish with Hyvän mielen pelit ry's skills?▼
You can structure multi-file implementation plans, delegate work to tiered subagents, budget Claude usage quotas per plan step, run PowerShell safely on Windows, write secret-shaped fixtures that survive GitHub push protection, and configure skill discovery for Claude Code, Codex, and Gemini Antigravity.
Who are these skills designed for?▼
They target engineers operating coding agents across Claude Code, Codex, and Gemini Antigravity who need shared conventions for planning, subagent delegation, and document reporting. Maintainers of shared plan repositories and Windows-based agent operators benefit most.
How do the skills get loaded and used at runtime?▼
Skills are discovered via .claude/skills pointer stubs for Claude Code or skills.json for Antigravity, with global configuration in ~/.gemini/config/skills.json. Most load contextually when planning or implementing; the quota-budget skill loads only on explicit user request.
What prerequisites do these skills require?▼
Prerequisites include a shared plans repository with an allowed-organization list, the additionalDirectories setting or /add-dir for Claude Code plan copying, the claude-budget.js script for quota tracking, and PowerShell 5.1 awareness for Windows command execution.
How do the skills handle credentials and secrets in code?▼
The agent-secret-hygiene skill defines what makes a string secret-shaped, prescribes two-concatenated literals for realistic test fixtures, lists safe placeholder forms, and gives recovery steps when GitHub push protection rejects a commit, explicitly warning against the bypass link.