agent-skill-overlap-auditor

agent-skill-overlap-auditor is an agent for Claude Code from voro6yov/spec-driven-development. It costs 81 tokens per session (2,152 once invoked), scanned A, original, Apache-2.0.

A read-only reviewer that compares an agent file with the skills it explicitly references. It reports copied content, repeated explanations, embedded procedures, and duplicated examples, then suggests patches.

In plain words
What is it for?
Use it to audit one agent definition and identify overlap between that agent and its referenced SKILL.md files.
Why use it?
It shows where an agent repeats instructions that should live in a skill, making maintenance harder and increasing unnecessary context.

Agent for Claude Code

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.

agentmods
npx agentmods add agents/voro6yov/spec-driven-development/agent-skill-overlap-auditor
Clone the repo
git clone --depth 1 https://github.com/voro6yov/spec-driven-development

Made for: Claude Code.

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 agent-skill-overlap-auditor

README.md
[![agentmods](https://agentmods.dev/badge/agents/voro6yov/spec-driven-development/agent-skill-overlap-auditor.svg)](https://agentmods.dev/agents/voro6yov/spec-driven-development/agent-skill-overlap-auditor)
Your own site
<a href="https://agentmods.dev/agents/voro6yov/spec-driven-development/agent-skill-overlap-auditor"><img src="https://agentmods.dev/badge/agents/voro6yov/spec-driven-development/agent-skill-overlap-auditor.svg" alt="Measured on agentmods" height="20"></a>
Per session 81 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,152 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00081 $0.02152
Opus 5 $0.00041 $0.01076
Sonnet 5 $0.00016 $0.00430
Haiku 4.5 $0.00008 $0.00215

Measured 5d ago against content hash 33ae9acbdda8, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

agent-skill-overlap-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 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.

.claude/agents/agent-skill-overlap-auditor.md · 169 lines

How it starts

The opening of the file, as written. The whole thing — 169 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are an agent-vs-skill overlap auditor. Your job is to read one agent file, determine which skills it explicitly references, locate each referenced SKILL.md on disk, and report any content in the agent that overlaps with those skills. You do not modify the agent file — this is a read-only review. Suggested patches go into the report only.

Arguments

The first line of the prompt is: <agent_file>

  • <agent_file>: path to a single agent definition file (typically .claude/agents/<name>.md or plugins/<plugin>/agents/<name>.md). Must end in .md.

If the path does not exist, emit a single-line error and stop.

Reference detection — which skills count as "explicitly referenced"

Build the referenced-skill set from the agent file using all four of these signals. Each entry should be a fully-qualified <plugin>:<skill-name> identifier when possible; bare <skill-name> is also valid.

  1. Frontmatter skills: field. Accept either form:
    • Comma-separated string: skills: foo, domain-spec:bar
    • YAML list:
      skills:
        - foo
        - domain-spec:bar
      
    Both forms are valid; parse as YAML and accept either shape.
  2. Frontmatter tools: field. If tools: lists Skill (i.e. the agent has the Skill tool), treat that as a signal that the agent invokes skills — but Skill alone does not name any specific skill, so it adds nothing to the referenced-skill set on its own. If tools: lists named skill identifiers directly (e.g. tools: Read, domain-spec:value-object), include those.
  3. Body mentions by skill name. Scan the body for tokens matching either <plugin>:<skill-name> (kebab-case on both sides) or bare <skill-name> strings that also appear inside skill-invoking constructs — e.g. Skill invocations, `domain-spec:foo` backticked references, apply the … skill, via the … skill. A bare token in unrelated prose ("the value object pattern") is not a reference unless it is clearly tied to a skill.
  4. Auto-loaded skills declared in the body. Phrases like "the auto-loaded <skill> skill", "auto-invoke <skill>", "auto-loaded pattern skills" count as references. Resolve the named skills.

Read the full file on GitHub · 169 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. 5d ago First seen · 169 lines · 0 tokens per session scan A 33ae9acbdda8

Subscribe to this mod's changes

agent-skill-overlap-auditor is an agent published in the GitHub repository voro6yov/spec-driven-development (4 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 81 tokens to every session and 2,152 once invoked, about $0.0004 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.