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 skills add marcusgoll/Spec-Flow --skill anti-duplicationgit clone --depth 1 https://github.com/marcusgoll/Spec-FlowWrote 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/skills/marcusgoll/spec-flow/anti-duplication)<a href="https://agentmods.dev/skills/marcusgoll/spec-flow/anti-duplication"><img src="https://agentmods.dev/badge/skills/marcusgoll/spec-flow/anti-duplication/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/marcusgoll/spec-flow/anti-duplication"><img src="https://agentmods.dev/badge/skills/marcusgoll/spec-flow/anti-duplication.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00056 | $0.07663 |
| Opus 5 | $0.00028 | $0.03832 |
| Sonnet 5 | $0.00011 | $0.01533 |
| Haiku 4.5 | $0.00006 | $0.00766 |
Grade A, and why
anti-duplication 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 9d 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 — 955 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Code duplication creates maintenance nightmares:
- Bug fixes must be applied in multiple places
- Features evolve inconsistently across duplicates
- Refactoring becomes exponentially harder
- Codebase grows unnecessarily large
- Onboarding takes longer (more code to learn)
This skill transforms "implement X" requests into a two-phase workflow:
- Search phase: Find existing similar implementations (endpoints, components, services, models)
- Reuse or justify: Either reuse/extend existing code OR justify why new implementation is necessary
The result: Codebases stay DRY, maintainable, and consistent.
<quick_start> <trigger_pattern> When user says "create", "implement", "add", or "build" new functionality, IMMEDIATELY search for existing similar code before writing anything new.
Trigger phrases:
- "Create a new API endpoint for..."
- "Implement a component that..."
- "Add a service to handle..."
- "Build a model for..."
- "Write a function that..." </trigger_pattern>
<basic_workflow> Step 1: Extract search keywords from request
- User: "Create an API endpoint to fetch user profile"
- Keywords: "API endpoint", "fetch", "user", "profile", "GET"
Step 2: Search codebase for similar patterns
PRIMARY: Semantic search (mgrep)
- mgrep for: "API endpoints that fetch user data", "profile retrieval handlers"
- Finds similar implementations by meaning, not exact text
SECONDARY: Literal search (Grep/Glob)
- Grep for: "router.get", "app.get", "/api/user", "profile"
- Glob for: "/user.ts", "/profile.ts", "/routes/"
Step 3: Analyze findings
- Found 3 similar endpoints: GET /api/user/:id, GET /api/user/settings, GET /api/admin/users
- All use same auth middleware, validation pattern, error handling
Step 4: Present reuse options
- Option A: Extend existing GET /api/user/:id to include profile data
- Option B: Create new endpoint using same patterns as existing user endpoints
- Option C: Justify why completely new approach is needed
Step 5: Implement chosen option (with existing patterns as template) </basic_workflow>
<immediate_value> Before anti-duplication skill: User: "Create an API endpoint to update user email" Claude: Writes new endpoint from scratch with different auth, validation, error handling Result: 5th variation of similar endpoint, inconsistent patterns
After anti-duplication skill: User: "Create an API endpoint to update user email" Claude: Searches codebase, finds 4 existing user endpoints with consistent patterns Claude: "Found 4 similar user endpoints. They all use: authMiddleware, validateUserInput, standardErrorHandler. I'll create the new endpoint following this established pattern." Result: Consistent implementation, reused middleware, maintainable code </immediate_value> </quick_start>
Extract key information:
- Type: What are they building? (endpoint, component, service, model, function, utility)
- Domain: What business domain? (user, product, order, payment, auth)
- Operation: What action? (create, read, update, delete, fetch, validate, transform)
- Technology: What stack? (React, Express, TypeScript, SQL, GraphQL)
Example parsing:
Request: "Create a React component to display product details"
Parsed:
- Type: React component
- Domain: Product
- Operation: Display/render
- Technology: React, TypeScript (likely)
Create multiple search approaches to maximize coverage:
PRIMARY: Semantic search (mgrep) Use mgrep FIRST for pattern discovery - finds similar code by meaning:
- "components that display product information"
- "services that handle email sending"
- "validation logic for user input"
- "error handling patterns in API routes"
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 9d ago First seen · 955 lines · 56 tokens per session scan A d5f82dc10cb6
anti-duplication is a skill published in the GitHub repository marcusgoll/Spec-Flow (92 stars, last pushed 4mo ago), licensed MIT. It adds 56 tokens to every session and 7,663 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 2026-08-30.
Other skills, from other repositories
android-ui-journey-testing
XML-specified Android UI journey testing, interactive step execution, assertion verification, and JSON outcome reporting.
agent-self-scheduling
Schedule AI agent runs with cron, loops, or external clocks while avoiding unsafe tight autonomous timers.
clawd-animation-lite
A lightweight generator for short, 1–3 second pixel animations featuring Clawd, the Claude Code mascot, shown as a small coral-orange crab. It creates a complete HTML file that can be opened in a browser.
clawd-animation
A generator for self-contained HTML pixel animations featuring Clawd, the pixel-style crab mascot of Claude Code. It turns a natural-language scene into a short looping browser animation with a background, character actions, props, and effects.
self-improving-agent
Low-noise self-improvement skill. Captures learnings, errors, and corrections into a local .learnings/ cache layer. Never auto-promotes to long-term memory files (SOUL.md, MEMORY.md, AGENTS.md, TOOLS.md). Promotion requires explicit user approval after repeated validation.
HTMLHero.skill
A workflow for building a working, previewable first-page hero section—a prominent opening area of a website—from a prompt, brand direction, and visual requirements. It can work with HTML or React projects and treats the hero as the visual template for later pages.