eval

eval is a skill for Claude Code from vocalbridgeai/vocal-bridge-claude-plugin. It costs 60 tokens per session (767 once invoked), scanned A, original, Apache-2.0.

A quality review of a recorded Vocal Bridge voice-agent call using the recording, transcript, tool activity, prompt, and session report.

In plain words
What is it for?
Use it to assess a call against an objective or scenario and receive a score, verdict, and improvement recommendations.
Why use it?
It identifies how well the agent handled a call and turns problems into specific suggestions for improving its instructions.

Skill for Claude Code

Written for Claude Code: allowed-tools in frontmatter.

Part of the vocal-bridge plugin — 16 skills shipped together

Good fit Use it to assess a call against an objective or scenario and receive a score, verdict, and improvement recommendations.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/vocalbridgeai/vocal-bridge-claude-plugin/eval
Install

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.

Any agent
npx skills add vocalbridgeai/vocal-bridge-claude-plugin --skill eval
Clone the repo
git clone --depth 1 https://github.com/vocalbridgeai/vocal-bridge-claude-plugin

Made for: Claude Code.

Or install vocal-bridge, the plugin that ships this one along with the rest of its 16 skills.

Wrote 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.

agentmods badge for eval

README.md
[![agentmods](https://agentmods.dev/badge/skills/vocalbridgeai/vocal-bridge-claude-plugin/eval.svg)](https://agentmods.dev/skills/vocalbridgeai/vocal-bridge-claude-plugin/eval)
Your own site
<a href="https://agentmods.dev/skills/vocalbridgeai/vocal-bridge-claude-plugin/eval"><img src="https://agentmods.dev/badge/skills/vocalbridgeai/vocal-bridge-claude-plugin/eval.svg" alt="Measured on agentmods" height="20"></a>
Per session 60 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 767 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00060 $0.00767
Opus 5 $0.00030 $0.00383
Sonnet 5 $0.00012 $0.00153
Haiku 4.5 $0.00006 $0.00077

Measured 7d ago against content hash 7eae27b22647, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, from the pricing page.

Security

Grade A, and why

eval 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 7d 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.

skills/eval/SKILL.md · 96 lines

How it starts

The opening of the file, as written. The whole thing — 96 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Evaluate a Vocal Bridge call recording with a multimodal LLM. Returns a structured score, verdict, and concrete suggestions for improving the agent's prompt.

First ensure CLI is installed and up to date:

pip install --upgrade vocal-bridge

Usage

vb eval <session_id> [--objective "..."] [--scenario "..."] [--objective-file FILE] [--scenario-file FILE] [--json]

The evaluator model is fixed by the platform.

Examples

# Basic eval against the agent's own system prompt
vb eval 550e8400-e29b-41d4-a716-446655440000

# With a specific objective
vb eval <session_id> --objective "Schedule an interview for next Tuesday"

# With both an objective and an expected scenario
vb eval <session_id> \
  --objective "Confirm the candidate's availability" \
  --scenario "The user is busy and tries to reschedule twice"

# Read long objective/scenario from files
vb eval <session_id> --objective-file objective.txt --scenario-file scenario.txt

# Raw JSON output (for piping into another tool)
vb eval <session_id> --json

What gets sent to the evaluator

  • Full audio recording of the call (inline, up to ~18 MB)
  • Agent's current system prompt (from vb config show)
  • Caller-supplied --objective and --scenario (optional)
  • Structured transcript with the agent's tool calls
  • Client action events log (heartbeats and custom bidirectional events between the agent and the client app)
  • Raw session report (for troubleshooting context)

The evaluator uses the audio as the primary source of truth for tone, interruptions, and timing, and uses the structured logs to verify what tools/actions actually fired.

Output

Call Evaluation
----------------------------------------
  Session:      550e8400...
  Score:        7/10
  Verdict:      partial

Summary:
  The agent answered the user's questions accurately but missed
  the scheduling objective when the user asked to reschedule.

What worked:
  + Greeted the caller naturally
  + Recovered cleanly from a mid-sentence interruption

What didn't:
  - Did not call schedule_meeting tool when the user gave a date
  - Tone became impatient on the second reschedule attempt

Suggested prompt improvements:
  Add an explicit instruction to call schedule_meeting whenever
  the user proposes any time, including reschedules.

Read the full file on GitHub · 96 lines

Changes

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.

  1. 7d ago First seen · 96 lines · 60 tokens per session scan A 7eae27b22647

Subscribe to this mod's changes

eval is a skill published in the GitHub repository vocalbridgeai/vocal-bridge-claude-plugin (7 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 60 tokens to every session and 767 once invoked, about $0.0003 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 2026-08-31.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

local-ai-agents

Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…

microsoft/ai-agents-for-beginners · 200 tokens

next-cache-components-adoption

Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…

vercel/next.js · 95 tokens

next-partial-prefetching-adoption

Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…

vercel/next.js · 103 tokens

chronicle

Analyze Copilot session history for standup reports, usage tips, session search, and session reindexing. Use when the user asks for a standup, daily summary, usage tips, workflow recommendations, wants to search or find past sessions by keyword/file/PR, wants to reindex their session store, or asks about deleting…

microsoft/vscode · 72 tokens

babysit-pr

Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…

openai/codex · 114 tokens