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 gyanranjan/polyagent-skills --skill implementation-sketchgit clone --depth 1 https://github.com/gyanranjan/polyagent-skillsWrote 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/gyanranjan/polyagent-skills/implementation-sketch)<a href="https://agentmods.dev/skills/gyanranjan/polyagent-skills/implementation-sketch"><img src="https://agentmods.dev/badge/skills/gyanranjan/polyagent-skills/implementation-sketch.svg" alt="Measured on agentmods" 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 | $0.00067 | $0.01720 |
| Opus 5 | $0.00034 | $0.00860 |
| Sonnet 5 | $0.00013 | $0.00344 |
| Haiku 4.5 | $0.00007 | $0.00172 |
Grade A, and why
implementation-sketch 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 8d 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 — 214 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Implementation Sketch
Purpose
Transform requirements or feature descriptions into actionable implementation plans with technical decisions, component breakdowns, task lists, and risk identification. The output should be sufficient for a development team to start building.
When to Use
- User has requirements and wants a technical plan
- User asks "how would we build this?"
- User wants to break a feature into development tasks
- User asks for a system design or architecture sketch
- User wants to evaluate technical approaches for a problem
When NOT to Use
- User needs requirements first (use
requirement-study) - User wants actual code written (that's coding, not sketching)
- User wants a presentation of the plan (use
deck-creatorafter this)
Inputs
Required:
- Requirements or feature description — what needs to be built
Optional:
- Tech stack constraints — languages, frameworks, infrastructure
- Team context — team size, skill sets, timeline
- Existing architecture — current system to extend
- Non-functional requirements — performance, scale, security targets
Process
Step 1: Preflight Tooling Check
Run a non-blocking Mermaid check before generating architecture diagrams:
if command -v mmdc >/dev/null 2>&1; then
mmdc --version
else
echo "mmdc not installed; continue with text diagram descriptions and add TODO entry"
fi
If mmdc is missing, update agent.todo.md with a tooling task.
Interaction Protocol (Mandatory)
For all substantive responses while running this skill:
- Include
Stage: Gx <name>andNext: <immediate next step> - At each gate transition, ask 2-4 decision-oriented questions before advancing
- Challenge weak assumptions and provide better alternatives where tradeoffs are clear
Step 2: Understand What's Being Built
Read the requirements/description thoroughly. Identify:
- Core functionality (what MUST work)
- Integration points (what connects to what)
- Data flows (what data moves where)
- User interactions (who does what)
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.
- 8d ago First seen · 214 lines · 67 tokens per session scan A 126ed571577b
implementation-sketch is a skill published in the GitHub repository gyanranjan/polyagent-skills (2 stars, last pushed 6mo ago), licensed MIT. It adds 67 tokens to every session and 1,720 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
architecture-paradigms
Selects and routes to the right architecture paradigm. Use when choosing patterns for a new system or comparing trade-offs before making architecture decisions.
implementing-architecture
Use when: about to write code and docs/architecture/ exists with STATUS: APPROVED in architecture.yaml. Not when: no approved architecture yet (use compiling-architecture first), or architecture.yaml lacks STATUS: APPROVED header.
task-decomposition
Use this skill to break down complex feature specifications, PRDs, or architectural designs into small, atomic, independently verifiable tasks and tickets. Ensures each task has clear acceptance criteria, dependencies, and validation steps.
architecture-diagram
Dark-themed SVG architecture/cloud/infra diagrams as HTML.
studio
Architecture Studio control plane — initialize or inspect a studio workspace, create and register projects, or route an architecture/AEC task to the right agent or skill. Use when the user runs /as:studio, asks to set up or open their studio, manage its projects, or describes a task without naming a skill.
project-execution
Executes implementation plans with progress tracking, checkpoint validation, and quality gates. Use after planning is complete and tasks are ready to implement.