Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add zajalist/hayba --skill hayba-save-load-regressiongit clone --depth 1 https://github.com/zajalist/haybaWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/zajalist/hayba/hayba-save-load-regression)<a href="https://agentmods.dev/skills/zajalist/hayba/hayba-save-load-regression"><img src="https://agentmods.dev/badge/skills/zajalist/hayba/hayba-save-load-regression/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/zajalist/hayba/hayba-save-load-regression"><img src="https://agentmods.dev/badge/skills/zajalist/hayba/hayba-save-load-regression.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00028 | $0.00367 |
| Opus 5.5 | $0.00011 | $0.00147 |
| Sonnet 5.5 | $0.00006 | $0.00073 |
| Haiku 4.5 | $0.00003 | $0.00037 |
Grade A, and why
hayba-save-load-regression scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 13d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
Test one game-save round trip
When to use
Use for one user-visible state such as checkpoint, inventory item, or objective completion. Identify the project's save slot and UI or input path, a disposable test profile, and the exact value expected after reload. Hayba has no generic SaveGame editor; do not invent a save operation.
Tool routing
- Search test_list for a project persistence test, run a focused candidate with test_run, and inspect test_get_log. If the game exposes a known save/load action, use editor_get_state before editor_start_pie, then trigger the action with editor_pie_press_key or editor_pie_click_widget after confirming it in editor_pie_widget_tree.
- Observe the state with editor_pie_actor_inspect or one-shot editor_pie_assert. End the session with editor_stop_pie. Start a fresh PIE session, invoke the game's load path, and observe the same state again. Poll editor_pie_wait_for in separate calls only if restoration is asynchronous; a single polling response is not a wait.
- Use an isolated test slot. If the save path or state is not visible through Hayba's tools, run the project's automated test or inspect it manually in Unreal and record that limitation. Never test against a player's real save without authorization.
Evidence
Report slot identity, pre-save state, save action and confirmation, fresh-session loaded state, test results, warnings, and unverified fields. A value that survives within one PIE session does not prove disk persistence or behavior after a process restart.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 13d ago First seen · 24 lines · 28 tokens per session scan A 9a6e89c663ad
hayba-save-load-regression is a skill published in the GitHub repository zajalist/hayba (13 stars, last pushed 3d ago), licensed MIT. It adds 28 tokens to every session and 367 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-04.
Other skills, from other repositories
pn-game-logic
Guides game loops, state machines, input handling, collision, scoring, and save state. Use when implementing game logic; covers architecture patterns for 2D/3D games.
unreal-playtest-delivery
Use when Unreal Engine work must be compiled, exercised in PIE, saved and reopened, packaged, launched, smoke-tested, or triaged from logs before a completion or delivery claim.
V3 Core Implementation
Core module implementation for claude-flow v3. Implements DDD domains, clean architecture patterns, dependency injection, and modular TypeScript codebase with comprehensive testing.
team-polish
Orchestrate the polish team — performance-analyst, technical-artist, sound-designer, qa-tester — to optimize and harden a feature or area.
work
Deliver one maintainer-approved EmDash issue, choosing the bug-fix path for a defect and the direct implementation path for an enhancement or task.
unity-agent-workflows
Use for AI-assisted Unity work that needs live repo discovery, project-derived routing, runtime-owner proof, runtime-visible output hard stops, runtime numeric proof for repeated visible-output failures, state-step guards, multi-agent scope ownership, modular C#/asmdef safety, UI/scene/visual asset gates, data-first…