topology-drift

topology-drift is a skill for Claude Code from dwarvesf/dwarves-kit. It costs 250 tokens per session (1,987 once invoked), scanned A, original, MIT.

A maintainer-only audit tool for checking whether a coding kit’s feature inventory matches its workflow map. It compares the documented commands, agents, skills, and hooks with the paths that describe how those features are used.

In plain words
What is it for?
Use it in the kit’s own repository to inventory features, compare them with the workflow-path index, and prepare fixes for differences.
Why use it?
It finds documentation drift after features are added or removed, so the workflow map does not quietly become incomplete or misleading.

Skill for Claude Code

Written for Claude Code: disable-model-invocation in frontmatter. Also seen: mentions subagents.

Needs its repository: it runs a file that does not travel with it, so clone the repository first. The line is bash lib/registry/feature-registry.sh generate /tmp/feat-check.md.

Part of the kit plugin — 11 skills, 38 commands, 30 agents, 10 hooks shipped together

Good fit Use it in the kit’s own repository to inventory features, compare them with the workflow-path index, and prepare fixes for differences.

Compare 6 skills from other repositories ↓
Install

Getting it into your agent

It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.

Clone the repo
git clone --depth 1 https://github.com/dwarvesf/dwarves-kit
agentmods
npx agentmods add skills/dwarvesf/dwarves-kit/topology-drift

Made for: Claude Code.

Or install kit, the plugin that ships this one along with the rest of its 11 skills, 38 commands, 30 agents, 10 hooks.

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 topology-drift

README.md
[![agentmods](https://agentmods.dev/badge/skills/dwarvesf/dwarves-kit/topology-drift.svg)](https://agentmods.dev/skills/dwarvesf/dwarves-kit/topology-drift)
Your own site
<a href="https://agentmods.dev/skills/dwarvesf/dwarves-kit/topology-drift"><img src="https://agentmods.dev/badge/skills/dwarvesf/dwarves-kit/topology-drift.svg" alt="Measured on agentmods" height="20"></a>
Per session 250 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,987 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. Third-party audits
  • NVIDIA SkillSpector pass 7 Sept 2026
How audits are shown
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 $0.00250 $0.01987
Opus 5 $0.00125 $0.00993
Sonnet 5 $0.00050 $0.00397
Haiku 4.5 $0.00025 $0.00199

Measured 4d ago against content hash e9b9952527af, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, from the pricing page.

Security

Grade A, and why

topology-drift 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 4d 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/topology-drift/SKILL.md · 77 lines

How it starts

The opening of the file, as written. The whole thing — 77 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Topology drift

Maintainer-only, dwarves-kit repo development only: this skill audits the KIT'S OWN docs/FEATURES.md/docs/workflow-paths.md pair, which only exists in this repo. It is not something a consumer/adopter repo would ever have a reason to invoke. To inventory the features of a project the kit is POINTED AT (a migration source, a codebase you want documented), use /kit:feature-map instead -- see docs/patterns/audit-loop.md for how the two relate.

Overview

Audit the live feature estate (every command, agent, skill, hook) against the workflow path map and ship fixes as a PR. This is the feature-inventory instance of docs/patterns/audit-loop.md: the generated registry docs/FEATURES.md (SPEC-219) is the machine truth; docs/workflow-paths.md section 5 is the hand-derived path index that drifts when features land or leave. Tier 1 is pure shell; Tier 2 spends model time only on the delta.

The four slots (per the audit-loop pattern)

Slot This instance
Item set every row of docs/FEATURES.md plus every line of docs/workflow-paths.md section 5 (both sides enumerate; the diff is the queue)
Contract (a) FEATURES.md is fresh (regenerating changes nothing); (b) every registry row has exactly one path-index line and vice versa, all four kinds; (c) trigger classes agree between the two (secondary, listed not auto-fixed)
Evidence class Tier 1: the repo itself via the generator diff + sed/comm name extraction. Tier 2: a model re-reads only the delta features' definitions to place them on the topology
Apply mechanics FIX on an isolated branch; new features get a path-index line AND a placement on the section 2/3 diagrams; removed features get their line deleted; PR body lists every verdict; UNSURE never auto-resolved

Process

  1. Freshness gate, REFUSE if stale. Regenerate to a temp file and compare:

    bash lib/registry/feature-registry.sh generate /tmp/feat-check.md
    cmp -s /tmp/feat-check.md docs/FEATURES.md || echo STALE
    

Read the full file on GitHub · 77 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. 4d ago First seen · 77 lines · 250 tokens per session scan A e9b9952527af

Subscribe to this mod's changes

topology-drift is a skill published in the GitHub repository dwarvesf/dwarves-kit (11 stars, last pushed today), licensed MIT. It adds 250 tokens to every session and 1,987 once invoked, about $0.0013 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 2026-09-04.