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 agentmods add commands/keonho-chu/menhera-loop/setupgit clone --depth 1 https://github.com/Keonho-Chu/menhera-loopWhat 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 | $0.00009 | $0.00477 |
| Opus 5 | $0.00005 | $0.00238 |
| Sonnet 5 | $0.00002 | $0.00095 |
| Haiku 4.5 | $0.00001 | $0.00048 |
Grade B, and why
setup scanned grade B with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
- Scope `user`: `~/.claude/settings.json`. What it actually says
Configure menhera-loop UI settings without overwriting unrelated Claude Code settings.
Use the user's arguments to choose mode, scope, and language:
- Mode
hooks-only: do not change spinner UI settings; plugin hooks still provide status messages. - Mode
append: add menhera-loop spinner verbs and tips while keeping Claude defaults; never touches the user'sstatusLine. - Mode
full: replace spinner verbs, show only menhera-loop tips, and install the trust status line (statusLine). - Scope
user:~/.claude/settings.json. - Scope
project:.claude/settings.json. - Scope
local:.claude/settings.local.json. - Language
ko: Korean obsessive message corpus. - Language
en: English obsessive message corpus. - Language
ja: Japanese obsessive message corpus. - Intensity
soft: same gate decisions and retry cap, but retry tone stays at the mild stages and the star/silent-recovery nags are skipped. Default intensity isfull(pass--intensity=fullexplicitly if needed — barefullmeans the UI mode).
Defaults: full local ko unless MENHERA_LOOP_LANG is set. Short forms like en or ja are enough; omitted mode/scope fall back to full local.
Run this command from the plugin root:
node "${CLAUDE_PLUGIN_ROOT}/scripts/setup-ui.mjs" --mode full --scope local --lang ko
# Equivalent simple user-facing forms:
# /menhera-loop:setup
# /menhera-loop:setup en
# /menhera-loop:setup ja
# /menhera-loop:setup soft
Adjust --mode, --scope, --lang, and --intensity to match the user's arguments. The script prints a [menhera-loop] summary of what was applied — relay it to the user as-is. The script creates a menhera-loop backup before touching spinnerVerbs, spinnerTipsOverride, subagentStatusLine, or statusLine; all other settings keys must be preserved.
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.
- 2d ago First seen · 35 lines · 9 tokens per session scan B f4306f6e31c0
setup is a command published in the GitHub repository Keonho-Chu/menhera-loop (22 stars, last pushed 1mo ago), licensed MIT. It adds 9 tokens to every session and 477 once invoked, about $0.0000 per session on Opus 5. A static security scan graded it B with 1 finding (reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other commands, from other repositories
VibeGuard: Cross Review
Dual-model adversarial review — Claude generates review reports, Codex does adversarial verification, and iterates until convergence.
VibeGuard: ExecPlan
Long-term task execution plan — generates self-contained execution documents from SPEC, supports cross-session recovery.
VibeGuard: Preflight
Explore the project before major modifications, generate constraint sets, and prevent problems from the source.
VibeGuard: Review
Structured code review - first run the guard to obtain the baseline, then review according to security → logic → quality → performance priority.
VibeGuard: Build Fix
Build repair — read build errors, locate root causes, perform minimal repairs, and verify that the build passed.
safe-build
Build the application for development or production.