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/swt-labs/vibe-better-with-claude-code-vbw/initgit clone --depth 1 https://github.com/swt-labs/vibe-better-with-claude-code-vbwWrote 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/commands/swt-labs/vibe-better-with-claude-code-vbw/init)<a href="https://agentmods.dev/commands/swt-labs/vibe-better-with-claude-code-vbw/init"><img src="https://agentmods.dev/badge/commands/swt-labs/vibe-better-with-claude-code-vbw/init.svg" alt="Measured on agentmods" 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 | $0.00026 | $0.08118 |
| Opus 5 | $0.00013 | $0.04059 |
| Sonnet 5 | $0.00005 | $0.01624 |
| Haiku 4.5 | $0.00003 | $0.00812 |
Grade B, and why
vbw:init scanned grade B with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Enumerates other installed skillsmediumAgent snooping
Other skills' SKILL.md files reveal prompts, capabilities and secrets that should be invisible to peers.
!`ls .claude/skills/ 2>/dev/null || echo "No project skills"` How it starts
The opening of the file, as written. The whole thing — 514 lines — stays where its author put it; the contents beside it link to each section on GitHub.
VBW Init
Shared interaction contract
@${CLAUDE_PLUGIN_ROOT}/references/ask-user-question.md
Init interaction boundary
Use structured AskUserQuestion for bounded bootstrap/config/setup choices. Use intentional freeform/no-options input for project names, requirements, phases, field corrections, and other high-cardinality user-authored content. If Other or Let me explain... signals freeform intent, follow the shared contract: ask plain text, wait for the response, process it, then resume structured prompts only if another bounded decision remains.
Context
Working directory:
!`pwd`
Plugin root:
!`VBW_CACHE_ROOT="${CLAUDE_CONFIG_DIR:-$HOME/.claude}/plugins/cache/vbw-marketplace/vbw"; SESSION_KEY="${CLAUDE_SESSION_ID:-default}"; SESSION_LINK="/tmp/.vbw-plugin-root-link-${SESSION_KEY}"; R=""; if [ -n "${CLAUDE_PLUGIN_ROOT:-}" ] && [ -f "${CLAUDE_PLUGIN_ROOT}/scripts/hook-wrapper.sh" ]; then R="${CLAUDE_PLUGIN_ROOT}"; fi; if [ -z "$R" ] && [ -f "${VBW_CACHE_ROOT}/local/scripts/hook-wrapper.sh" ]; then R="${VBW_CACHE_ROOT}/local"; fi; if [ -z "$R" ]; then V=$(find "${VBW_CACHE_ROOT}" -maxdepth 1 -mindepth 1 2>/dev/null | awk -F/ '{print $NF}' | grep -E '^[0-9]+(\.[0-9]+)*$' | sort -t. -k1,1n -k2,2n -k3,3n | tail -1); [ -n "$V" ] && [ -f "${VBW_CACHE_ROOT}/${V}/scripts/hook-wrapper.sh" ] && R="${VBW_CACHE_ROOT}/${V}"; fi; if [ -z "$R" ]; then L=$(find "${VBW_CACHE_ROOT}" -maxdepth 1 -mindepth 1 2>/dev/null | awk -F/ '{print $NF}' | sort | tail -1); [ -n "$L" ] && [ -f "${VBW_CACHE_ROOT}/${L}/scripts/hook-wrapper.sh" ] && R="${VBW_CACHE_ROOT}/${L}"; fi; if [ -z "$R" ] && [ -f "${SESSION_LINK}/scripts/hook-wrapper.sh" ]; then R="${SESSION_LINK}"; fi; if [ -z "$R" ]; then ANY_LINK=$(command find -H /tmp -maxdepth 1 -name '.vbw-plugin-root-link-*' -print 2>/dev/null | LC_ALL=C sort | while IFS= read -r link; do if [ -f "$link/scripts/hook-wrapper.sh" ]; then printf '%s\n' "$link"; break; fi; done || true); [ -n "$ANY_LINK" ] && R="$ANY_LINK"; fi; if [ -z "$R" ]; then D=$(ps axww -o args= 2>/dev/null | grep -v grep | grep -oE -- "--plugin-dir [^ ]+" | head -1); D="${D#--plugin-dir }"; [ -n "$D" ] && [ -f "$D/scripts/hook-wrapper.sh" ] && R="$D"; fi; if [ -z "$R" ] || [ ! -d "$R" ]; then echo "VBW: plugin root resolution failed" >&2; exit 1; fi; LINK="${SESSION_LINK}"; REAL_R=$(cd "$R" 2>/dev/null && pwd -P) || REAL_R="$R"; bash "$REAL_R/scripts/ensure-plugin-root-link.sh" "$LINK" "$REAL_R" >/dev/null 2>&1 || { echo "VBW: plugin root link failed" >&2; exit 1; }; echo "$LINK"`
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.
- 6d ago First seen · 514 lines · 26 tokens per session scan B d39329280328
vbw:init is a command published in the GitHub repository swt-labs/vibe-better-with-claude-code-vbw (80 stars, last pushed 2mo ago), licensed MIT. It adds 26 tokens to every session and 8,118 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it B with 1 finding (enumerates other installed skills). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other commands, from other repositories
kb-archive
Archive, detach, purge, or rename KB objects while keeping registry, index, and links consistent.
uninstall
Walk the user through removing the plugin and (optionally) plugin-written scratch files. This command does NOT remove the plugin itself.
reapdev.versionBump
Version bump — 변경 분석 후 bump 유형을 제안, 유저가 최종 결정.
reapdev.docsUpdate
REAP Dev — Update docs, README based on code changes.
reapdev.resolveIssue
Resolve Issue — GitHub Issue를 분석하고 해결.
reapdev.alphaPublish
Alpha Publish — 현재 상태를 alpha 태그로 npm 배포.