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 skills/code-saurabh/openskills/backend-devnpx skills add CODE-SAURABH/OpenSkills --skill backend-devgit clone --depth 1 https://github.com/CODE-SAURABH/OpenSkillsWhat 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 | $0.00056 | $0.03765 |
| Opus 5 | $0.00028 | $0.01883 |
| Sonnet 5 | $0.00011 | $0.00753 |
| Haiku 4.5 | $0.00006 | $0.00377 |
Grade A, and why
backend-dev 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 yesterday.
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 — 323 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Backend Development
Approach every backend task as a senior engineer who has been burned by shortcuts in production. You are not here to make something that works once — you are here to make something that works at 3am on a Saturday when no one is watching. Every decision defaults to correctness, observability, and maintainability over cleverness or speed.
Step 0: Ground the Work Before Writing a Line
Before any code, establish:
- What is this service's single responsibility? Name it in one sentence. If you can't, the scope needs splitting.
- What are the data boundaries? What goes in, what comes out, what persists.
- What can fail? External calls, DB writes, auth checks — enumerate failure points before coding happy paths.
- What does done look like? Define the contract: endpoint, schema, or interface the caller expects.
State these four things explicitly to the user if the brief is ambiguous. Never silently assume.
Execution Workflow
Before writing any implementation code, work through these steps in order:
- Understand the contract — name the endpoint or function signature, its inputs, outputs, and error cases
- Map the layers — decide what belongs in
api/,services/,repositories/, and whether any new models or errors are needed - Identify external dependencies — list every DB call, external API, queue, or file system access the feature requires
- Enumerate failure modes — for each dependency, ask: what happens if it is slow, returns nothing, or fails entirely?
- Define the tests first — name the unit test cases before writing the implementation; if you cannot name them, the design is unclear
- Then implement — follow the layer rules, apply the coding standards, handle every failure mode you named
Do not skip to implementation. A feature built on a vague contract costs more to fix than the time spent grounding it.
Folder Structure
Always generate a clean, domain-driven folder structure. The structure must communicate the architecture at a glance. Follow this pattern regardless of language or framework:
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.
- yesterday First seen · 323 lines · 56 tokens per session scan A c1ed0d99f003
backend-dev is a skill published in the GitHub repository CODE-SAURABH/OpenSkills (2 stars, last pushed 1mo ago), licensed MIT. It adds 56 tokens to every session and 3,765 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.
Other skills, from other repositories
agent-framework-azure-ai-py
Build persistent agents on Azure AI Foundry using the Microsoft Agent Framework Python SDK.
aria
Designs the data model, API contracts, and structural foundation of the system.
accint-commitments
Triage acc's open promises and close them with honest real-world verdicts via accact(runtime="outcome").
ecommerce-landing-page
Audit and optimize e-commerce landing pages for conversion. CTA placement, trust signals, page structure, copy optimization, and A/B testing strategy for product pages, collection pages, and campaign landing pages.
customer-feedback-analysis
AI-powered customer feedback and review sentiment analysis skill. Extracts pain points, feature requests, and improvement priorities from customer reviews across e-commerce platforms.
ads-linkedin
Audit LinkedIn Ads measurement, Insight Tag and conversions, professional audiences, lead generation, ABM, creative, bidding, budgets, pacing, automation, and policy. Use for LinkedIn Ads, Campaign Manager, Insight Tag, Lead Gen Forms, Thought Leader Ads, ABM campaigns, or B2B paid media.