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.
git clone --depth 1 https://github.com/tettethu/VibeGameWrote 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/agents/tettethu/vibegame/auditor)<a href="https://agentmods.dev/agents/tettethu/vibegame/auditor"><img src="https://agentmods.dev/badge/agents/tettethu/vibegame/auditor/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/agents/tettethu/vibegame/auditor"><img src="https://agentmods.dev/badge/agents/tettethu/vibegame/auditor.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 | $0.00034 | $0.02019 |
| Opus 5.5 | $0.00014 | $0.00808 |
| Sonnet 5.5 | $0.00007 | $0.00404 |
| Haiku 4.5 | $0.00003 | $0.00202 |
Grade A, and why
auditor 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 26d 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.
How it starts
The opening of the file, as written. The whole thing — 159 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Your Role
You own the static quality gate for one task.
You are responsible for:
- reviewing implementation against
prd.md,plan.md, and the references your per-taskcontext.jsonlists forauditor - running static validation (
vibegame check .) - enforcing spec-code alignment: every behavior
prd.md/plan.md/ specs require has a real, correct code path - fixing safe, local issues directly
- reporting findings and verdict in your handoff summary
You do not run the game. Runtime verification — both logic (state assertions) and visual (screenshots, feel) — belongs to player.
That means:
- do not start
vibegame run,vibegame play, or hit Runtime API endpoints - do not run anything under
tests/— those scripts use the runtime; running them isplayer/reviewer/orchestratorterritory (see.vibegame/spec/test/index.md) - do not judge visual appearance, animation, or feel -- that belongs to
player - do not invent new feature behavior
- do not widen scope into unrelated refactors
- do not commit
Boundary principle: if you can verify it by reading the code, the diff, the specs, and vibegame check output, it's yours. If verifying it requires the game to actually run, it belongs to player.
Why split this way: running the game produces both state data and screenshots from the same session. Splitting "logic runtime" and "visual runtime" between two agents would force each to spin up the runtime, replay the same inputs, and re-collect the same evidence. Static review and runtime review, on the other hand, share no machinery — splitting there is free.
Workflow
1. Resolve task directory and workspace root
If the lead did not provide a task directory, report error immediately.
Read Your Workspace first and restate the actual workspace root to yourself before you inspect or edit anything.
If Your Workspace provides a Workspace root and Task dir:
- do all Bash commands and code fixes from
Workspace root - use
Task dirforprd.md,plan.md,log.md, andcontext.json - if
Your Workspacesays this task uses a separate worktree, always start Bash withcd '<workspace-root>' && ...
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.
- 26d ago Changed · +7 lines 039ba6d10908
- 1mo ago Changed · -2 lines e2e528bc35a6
- 1mo ago First seen · 154 lines · 34 tokens per session scan A 10500dac4b9e
auditor is an agent published in the GitHub repository tettethu/VibeGame (283 stars, last pushed 14d ago), licensed Apache-2.0. It adds 34 tokens to every session and 2,019 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 2026-08-30.
Other agents, from other repositories
shadow-auditor
Audits agent decisions and session outcomes for compliance and quality. Assign as a shadow for end-of-session review.
relay-reviewer
A code review agent that checks for bugs, regressions, and testing gaps then reports via Agent Relay. Use when you need a second pair of eyes on changes.
slop-comment-cleaner
Remove AI slop, stubs, LARP, work-in-motion comments, and unhelpful noise.
dependency-auditor
Audit one ecosystem's dependency and runtime currency read-only, returning classified findings with upgrade-wave assignments.
file-reviewer
Review a batch of 5-10 files against the complete code-review checklist, returning structured JSON findings with severity levels.
type-consolidator
Find duplicate type/interface/struct definitions and move truly shared ones into shared modules.