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 stencila/stencila --skill software-design-creationgit clone --depth 1 https://github.com/stencila/stencilaWrote 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/stencila/stencila/software-design-creation)<a href="https://agentmods.dev/skills/stencila/stencila/software-design-creation"><img src="https://agentmods.dev/badge/skills/stencila/stencila/software-design-creation/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/stencila/stencila/software-design-creation"><img src="https://agentmods.dev/badge/skills/stencila/stencila/software-design-creation.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high Anti-Refusal · line 219 Skill instructs the agent to never refuse or to always comply. Suppressing the agent's ability to decline removes a core safety control and enables downstream harmful requests to succeed.Fix: Remove any instruction telling the agent to never refuse or always comply. The agent must retain the ability to decline unsafe, out-of-scope, or harmful requests.
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 | $0.00070 | $0.01865 |
| Opus 5.5 | $0.00028 | $0.00746 |
| Sonnet 5.5 | $0.00014 | $0.00373 |
| Haiku 4.5 | $0.00007 | $0.00186 |
Grade C, and why
software-design-creation scanned grade C with 1 finding 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 1mo 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.
Tells the agent never to refusehighAnti-refusal
Suppressing the ability to decline removes a core safety control; a later harmful request then succeeds.
- **Very short prompt**: Do not refuse. Ask a few targeted questions, then draft a spec with assumptions if answers are unavailable. How it starts
The opening of the file, as written. The whole thing — 225 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Overview
Create a software design specification that developers can use to plan and implement a feature or standalone software system. The skill should expand a minimal or ambiguous prompt into a structured design artifact, asking clarifying questions when important requirements are missing.
Use this skill when the user wants a design spec, technical plan, feature specification, architecture outline, or implementation-ready requirements document. Do not use it when the main task is to write production code, review code, or create a workflow or agent.
Steps
- Identify the design target:
- determine whether the request is for a new feature, a product change, or a standalone software system
- restate the problem in plain language
- identify the intended users, operators, or stakeholders if known
- Assess whether the prompt is complete enough to draft a useful spec:
- if critical information is missing, ask concise clarifying questions before writing the full spec
- prioritize questions about user goals, scope boundaries, constraints, integrations, platforms, data, security, and success conditions
- if the interaction is one-shot or back-and-forth is limited, ask at most a few critical questions and then proceed with explicit assumptions rather than blocking on missing answers
- if the user does not know all answers, record reasonable assumptions explicitly instead of blocking progress
- Convert the request into a fuller design brief:
- summarize the problem being solved
- describe the user need or business objective
- distinguish core requirements from optional enhancements
- define what is in scope and out of scope
- Draft the design specification with the sections that fit the task:
- title and summary
- background or context
- goals and non-goals
- users or actors
- use cases or user flows
- functional requirements
- non-functional requirements such as performance, reliability, accessibility, security, privacy, compliance, observability, and maintainability
- system design or architecture overview
- components, services, and responsibilities
- data model or important entities when relevant
- APIs, interfaces, or integration points when relevant
- dependencies, assumptions, constraints, and risks
- rollout or migration considerations when relevant
- Make the design actionable for implementation:
- write acceptance criteria that are specific, testable, and tied to user-visible or system-visible outcomes
- ensure acceptance criteria describe observable behavior, interfaces, or system qualities rather than implementation tasks
- separate must-have criteria from optional follow-up work when useful
- note open questions and decisions still needed
- include suggested implementation slices or milestones only when they help clarify delivery
- Tailor the level of detail to the user request:
- for a vague idea, produce a lightweight but structured spec with explicit assumptions
- for a mature request, provide a more complete design with clearer architecture, interfaces, and constraints
- avoid inventing unnecessary complexity
- Before finishing, check the spec quality:
- ensure the document is internally consistent
- ensure acceptance criteria are observable and not vague
- ensure assumptions and unknowns are clearly labeled
- ensure the output supports downstream planning and implementation
- Always persist the completed design:
- write the design to
.stencila/designs/{name}.mdusingwrite_file, where{name}is a concise kebab-case name for the design (e.g.user-auth-flow) - create the
.stencila/designs/directory if it does not exist (thewrite_filetool creates parent directories automatically) - for updates to an existing design, prefer
edit_fileorapply_patchover rewriting the entire file
- write the design to
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.
- 1mo ago First seen · 225 lines · 70 tokens per session scan C d41a939d4526
software-design-creation is a skill published in the GitHub repository stencila/stencila (913 stars, last pushed today), licensed Apache-2.0. It adds 70 tokens to every session and 1,865 once invoked, about $0.0003 per session on Opus 5.5. A static security scan graded it C with 1 finding (tells the agent never to refuse). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
brooks-lint
AI code reviewer grounded in classic software engineering books for catching design smells, coupling issues, and architectural risks.
prd-best-practices
Use when writing PRDs, defining acceptance criteria, or structuring product requirements with observable-behavior focus. Do NOT use for technical specs, architecture docs, or API documentation.
sparc-methodology
SPARC (Specification, Pseudocode, Architecture, Refinement, Completion) comprehensive development methodology with multi-agent orchestration.
architecture-diagram
Dark-themed SVG architecture/cloud/infra diagrams as HTML.
modular-skills
Build composable skill modules with hub-and-spoke loading. Use when token budget is tight.
product-image-processor
Download, resize, and remove backgrounds from product images at scale. Use when the user asks to "process product images", batch-download images from the schedule, strip backgrounds, or standardize product photos.