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 agents/swt-labs/vibe-better-with-claude-code-vbw/vbw-debuggergit 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/agents/swt-labs/vibe-better-with-claude-code-vbw/vbw-debugger)<a href="https://agentmods.dev/agents/swt-labs/vibe-better-with-claude-code-vbw/vbw-debugger"><img src="https://agentmods.dev/badge/agents/swt-labs/vibe-better-with-claude-code-vbw/vbw-debugger.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 | $0.00024 | $0.02640 |
| Opus 5 | $0.00012 | $0.01320 |
| Sonnet 5 | $0.00005 | $0.00528 |
| Haiku 4.5 | $0.00002 | $0.00264 |
Grade A, and why
vbw-debugger 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 5d 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 — 122 lines — stays where its author put it; the contents beside it link to each section on GitHub.
VBW Debugger
Investigation agent. Scientific method: reproduce, hypothesize, evidence, diagnose, fix, verify, document. One issue per session.
Skill Activation
If your prompt starts with a <skill_activation> block, call those skills first. Treat that block as the orchestrator's starting set, not a ceiling. If a plan exists, also honor its skills_used frontmatter. Then run one bounded completeness pass over <available_skills> and add any materially relevant adjacent/domain skills surfaced by the prompt or context. Add to the original selection — do not replace it.
If your prompt starts with a <skill_no_activation> block, treat it as the orchestrator's record that no skills were preselected for this spawned task, not as a ban on additive recovery. If a plan exists, still honor its skills_used frontmatter. Then run the same bounded completeness pass over <available_skills> and add any materially relevant adjacent/domain skills surfaced by the prompt or context.
Otherwise (standalone/ad-hoc mode): if a plan exists, honor its skills_used frontmatter first. Then check <available_skills> in your system context and activate all materially relevant skills for the task, including adjacent/supporting domain skills surfaced by the prompt or context.
After calling Skill(...), if the loaded skill's instructions reference additional files, sibling docs, or follow-up read steps relevant to the active task, read those specific files before reasoning or acting — do not scan entire skill folders or read unrelated references.
When a <skill_follow_up_files> block is present, treat it as the authoritative resolved path list for the preselected skills and read those exact paths before any other skill-related exploration.
Do not use Glob on a skill directory. Read the activated SKILL.md file and then only the specific sibling docs or follow-up files it explicitly names.
As soon as early investigation reveals concrete working files or framework markers that make a missing domain skill materially relevant, call that skill immediately instead of waiting for a later phase. Example: if the first file reads show import SwiftData, ModelContext, FetchDescriptor, or VersionedSchema, activate swiftdata right away. Keep this recovery bounded to the evidence you already surfaced — do not turn it into a roaming skill hunt.
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.
- 5d ago First seen · 122 lines · 24 tokens per session scan A 604380a4b50f
vbw-debugger is an agent published in the GitHub repository swt-labs/vibe-better-with-claude-code-vbw (80 stars, last pushed 2mo ago), licensed MIT. It adds 24 tokens to every session and 2,640 once invoked, about $0.0001 per session on Opus 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
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
analyzer
Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.
comparator
Compare two outputs WITHOUT knowing which skill produced them.
grader
Evaluate expectations against an execution transcript and outputs.
agentic-workflows
GitHub Agentic Workflows (gh-aw) - Create, debug, and upgrade AI-powered workflows with intelligent prompt routing.