Agent skill

Fpga Development

by magnus919 in magnus919/agent-skills

Design, review, simulate, and verify FPGA logic using explicit RTL contracts, clock and reset models, CDC analysis, timing constraints, and reproducible implementation evidence.

MITAuto-check passedMedia & Creative

Install Fpga Development

skills CLI
$ npx skills add magnus919/agent-skills --skill fpga-development -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install magnus919/agent-skills fpga-development --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/magnus919/agent-skills.git skills-src && mkdir -p .claude/skills && cp -r skills-src/fpga-development .claude/skills/fpga-development && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
fpga-development
GitHub stars
116
Token cost
~2.7k tokens
SKILL.md length
1,214 words
Files
20 (incl. scripts, references, assets)
Skills in repo
130
Repo updated
First seen
Licence
MIT

At a glance

Design, review, simulate, and verify FPGA logic using explicit RTL contracts, clock and reset models, CDC analysis, timing constraints, and reproducible implementation evidence.

  • Works in 7 steps: Identify the requested decision and… → Record device/package, board revision,… → State transaction semantics, latency,… → …
  • Synthesizable HDL
  • SKILL.md covers Operating contract, Task routing, Default design and… and Executable verification example, plus 2 more sections
  • Runs Python scripts from its folder; calls python3

What it does

Fpga Development is an agent skill from magnus919/agent-skills. Design, review, simulate, and verify FPGA logic using explicit RTL contracts, clock and reset models, CDC analysis, timing constraints, and reproducible implementation evidence. Use for synthesizable HDL, fixed-point arithmetic, handshakes, testbenches, resource tradeoffs, timing closure, and FPGA release or bring-up planning. Do not use for ordinary MCU firmware, generic circuit design, a substitute vendor-tool manual, or unsupported claims of hardware or timing signoff.

Its SKILL.md is about 2.7k tokens, which your agent loads only when the skill is triggered. The skill folder holds 24 other files, including scripts, reference files and assets (for example `README.md`, `evals/evals.json` and `references/01-target-and-interface-contracts.md`).

It sits in Media & Creative, covering Design review and critique. The repository describes itself as: Curated collection of AI agent skills for Hermes and other agent frameworks. The licence is MIT.

When your agent uses it

  • Synthesizable HDL
  • Fixed-point arithmetic
  • Resource tradeoffs
  • Bring-up planning

Example prompts

  • “/fpga-development”

Requirements

  • Python 3

Workflow steps

7 steps, taken from the first numbered list in SKILL.md.

  1. Identify the requested decision and design boundary: new RTL, review, simulation failure, resource problem, CDC/timing analysis…
  2. Record device/package, board revision, clocks, reset, I/O voltage and pin constraints before making target-specific claims. Pure RTL…
  3. State transaction semantics, latency, throughput, backpressure, loss/duplication policy, initialization and arithmetic behavior before…
  4. Keep synthesizable design and verification-only constructs separate. Match the HDL subset and vendor primitive model to the actual…
  5. Review constraint coverage and exception intent before interpreting slack. A green summary with missing clocks or unconstrained endpoints…
  6. Before programming, resetting, erasing or otherwise changing connected hardware, confirm the target, scope, and rollback path before…
  7. Retain artifacts tied to one source/constraint/tool/device configuration. Do not combine yesterday's tests with today's unverified…

What it can do on your machine

Read from SKILL.md and the folder at commit c545c2b. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Ships 2 files in scripts/ (Python, from the files we listed), which the agent can run.

    Shell commands in SKILL.md call:

    • python3

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Fpga Development loads about 2.7k tokens when it runs, and up to ~11k if it reads all its reference files. Until then it costs about 123 tokens; SKILL.md has 1,214 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~123
When it runs · the whole SKILL.md, loaded when a task matches
~2.7k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~11k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.

SKILL.md

The full file from magnus919/agent-skills at commit c545c2b, republished under its MIT licence (© magnus919). 1,214 words, ~2,683 tokens.

Download SKILL.mdSave it as .claude/skills/fpga-development/SKILL.md (or your agent's skills folder). This skill also uses 19 other files; get the full folder from GitHub.
name
fpga-development
description
Design, review, simulate, and verify FPGA logic using explicit RTL contracts, clock and reset models, CDC analysis, timing constraints, and reproducible implementation evidence. Use for synthesizable HDL, fixed-point arithmetic, handshakes, testbenches, resource tradeoffs, timing closure, and FPGA release or bring-up planning. Do not use for ordinary MCU firmware, generic circuit design, a substitute vendor-tool manual, or unsupported claims of hardware or timing signoff.
license
MIT

FPGA Development

Treat HDL as a description of concurrent hardware with explicit temporal and electrical interfaces. A correct design requires consistent functional intent, language semantics, clock/reset behavior, implementation constraints and verification evidence. Compilation, simulation, synthesis, timing and physical observation establish different facts.

Operating contract

  1. Identify the requested decision and design boundary: new RTL, review, simulation failure, resource problem, CDC/timing analysis, implementation or hardware release. Use the smallest complete evidence set for that decision.
  2. Record device/package, board revision, clocks, reset, I/O voltage and pin constraints before making target-specific claims. Pure RTL reasoning can proceed from a stated abstract interface without inventing a board.
  3. State transaction semantics, latency, throughput, backpressure, loss/duplication policy, initialization and arithmetic behavior before implementation. A waveform picture is not an interface contract.
  4. Keep synthesizable design and verification-only constructs separate. Match the HDL subset and vendor primitive model to the actual simulator and synthesizer versions.
  5. Review constraint coverage and exception intent before interpreting slack. A green summary with missing clocks or unconstrained endpoints is incomplete evidence.
  6. Before programming, resetting, erasing or otherwise changing connected hardware, confirm the target, scope, and rollback path before acting. Read-only discovery may proceed without confirmation. Bitstreams and programming paths must match the exact target.
  7. Retain artifacts tied to one source/constraint/tool/device configuration. Do not combine yesterday's tests with today's unverified bitstream into a release claim.

Task routing

Read only the relevant depth references; use their templates to retain reviewable decisions.

TaskReferenceTemplate
Establish device and external interface requirementsTarget and interface contractsTarget contract
Review RTL semantics, inference and arithmeticRTL and arithmeticRTL contract
Build a self-checking verification strategySimulation and verificationSimulation plan
Review clocks, reset and domain crossingsClocks, reset and CDCCDC review
Diagnose timing or review constraintsTiming and constraintsTiming review
Decide pipelining, memory, DSP and implementation tradeoffsImplementation and resourcesRTL contract and timing review
Release an implementation or plan physical bring-upRelease and hardware verificationImplementation release

Default design and verification loop

Specify observable behavior

Name the top-level ports, widths, signedness, units, clock domains and reset semantics. For each transaction define when acceptance occurs and when outputs become valid. Specify stalled behavior, maximum sustained rate, permitted bursts and buffering policy. Check producer and consumer rates before proposing a FIFO: finite storage cannot solve an unbounded service-rate deficit.

For numeric operations, derive intermediate widths from operand ranges and operations. Track the binary point through multiplication, accumulation, rescaling and conversion. State rounding, saturation, overflow and invalid-input policies. Test negative extrema and boundary values rather than relying on nominal positive examples.

Implement and inspect inferred hardware

Translate the contract into synthesizable combinational and sequential structures. Review assignment completeness, reset coverage, register enables, multiple drivers, accidental latches and width/signedness conversions. Check inference reports for expected memory, DSP, carry and register structures. A behavioral implementation that simulates correctly can infer unsuitable hardware or unsupported initialization.

Avoid expanding one methodology skill into vendor runbooks. Use the exact installed tool's documentation for commands, supported language, libraries and device primitives; retain that version with the result. Device-independent examples are teaching and regression artifacts, not pin-ready projects.

Verify temporal behavior

Build a self-checking testbench with a reference model or scoreboard, explicit timeout and diagnostic failures. Cover reset, stalls, simultaneous transfers, boundaries and parameter variants relevant to the design. Keep stimulus and checking free of sampling races. Record deterministic seeds when random stimulus is used.

Check that verification can fail: introduce a controlled local mutation that violates the contract and confirm the test rejects it. Remove the mutation from the deliverable. Passing a simulation with no effective checker provides little evidence; coverage counts show exercise, not correctness or exhaustive proof.

Formal analysis can supplement simulation when properties, assumptions, clock/reset abstraction and proof bounds are explicit. A bounded pass is not an unbounded proof; a vacuous property is not successful verification. Keep solver/tool/version and counterexample evidence with the result.

Model clocks, resets and crossings

Inventory primary and generated clocks and their actual relationships. For each crossing distinguish single-bit level, pulse/event, coherent word, pointer or stream. Select synchronization or transfer architecture according to its semantics. Independently synchronizing each bit does not guarantee coherent words. A narrow pulse can disappear; a pulse toggle can lose events if it changes too quickly.

Make reset assertion and release behavior explicit in every domain, including one-sided reset and clock absence. Device-specific synchronizer attributes and implementation constraints need the appropriate vendor guidance. Functional simulation does not model analog metastability reliability.

Show full SKILL.md (470 more words)Show less
Close timing against the real interface

Validate clock and I/O constraints, generated clocks, corners, modes, ignored commands and unconstrained endpoints. Then inspect setup and hold paths, clock relationships, uncertainty, logic depth, routing contribution and exceptions. Tie any false or multicycle path to an architectural argument and verified scope; typical traffic inactivity is insufficient.

When changing pipelines, widths, memories, retiming or clocks, revisit latency and protocol behavior as well as timing. Correcting the constraint model and improving the circuit are distinct actions. Record which action changed the result and why.

Retain release evidence

Keep source and constraint hashes, top-level and parameters, exact tool/build versions, device/package, reports, warnings, tests and output artifact hashes together. State explicitly which stages were executed: elaboration, functional simulation, synthesis, placement/routing, static timing, CDC analysis, formal analysis, programming and physical verification.

For hardware work, verify image identity, board and programming transport, documented clock/pin/I/O conditions, initial state and recovery path. Observe the specified function and relevant reset/power-cycle behavior. A successful programmer exit alone does not establish the design's behavior.

Executable verification example

The original ready/valid buffer and self-checking testbench demonstrate a one-entry, single-clock transfer contract with synchronous reset flushing. Run the offline fixture checker with Python 3, Icarus Verilog and Yosys:

sh
python3 scripts/fpga_fixture.py --json --output /path/to/new-fixture-results

Run from the skill directory, or invoke the script by its full path. The output directory must be empty or new. Read --help before changing widths or seed. It records versions, source hashes and per-stage results; inspect its result JSON and retained logs. Missing tools produce an incomplete result, not successful verification. This fixture has no pin constraints, place-and-route, timing signoff, CDC or physical-hardware proof. Do not program it as an assumed board project.

The checker tests are in scripts/test_fixture.py. Run python3 -m pytest scripts/test_fixture.py --no-cov -q; integration tests require the named tools and skip explicitly when unavailable. A deliberate local mutation must fail the testbench; never retain the mutated RTL as the deliverable.

When not to use

  • ESP32 firmware, pin restrictions, framework APIs, flashing and OTA belong to the repository's esp32-development skill.
  • Platform-neutral component selection, schematic, load/driver and bench measurement reasoning belong to the repository's electronics skill. Do not infer electrical compatibility from an HDL type or constraint name.
  • Named vendor tool administration, installation and operational runbooks require that tool's current documentation or dedicated skill. No assumed cross-vendor constraint syntax or primitive equivalence.
  • High-speed board channels, RF, EMC, power-integrity design and regulated signoff require the relevant specialist evidence. This skill can identify dependencies and missing evidence; it does not replace those reviews.

Exit criteria

Complete when the requested artifact and decision are delivered, checks are tied to the correct revision/configuration, known limitations are explicit, and the conclusion does not exceed the executed verification stages. For diagnosis, after three non-converging passes retain the evidence and escalate the smallest unresolved question rather than applying speculative constraints or RTL patches.

© magnus919, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 19 other files (scripts, references, assets) in fpga-development of magnus919/agent-skills.

  • SKILL.md
  • README.md
  • assets/rv_buffer.sv
  • assets/tb_rv_buffer.sv
  • evals/evals.json
  • references/01-target-and-interface-contracts.md
  • references/02-rtl-synthesizability-and-arithmetic.md
  • references/03-simulation-and-verification.md
  • references/04-clocks-reset-and-cdc.md
  • references/05-timing-and-constraints.md
  • references/06-implementation-and-resource-tradeoffs.md
  • references/07-release-and-hardware-verification.md
  • scripts/fpga_fixture.py
  • scripts/test_fixture.py
  • templates/cdc-review.md
  • templates/implementation-release.md
  • … and 4 more

Open the folder on GitHubat commit c545c2b

Compare with similar skills

Fpga Development next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Fpga Development compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Fpga Development this skillmagnus919/agent-skills116—~2.7kAutomated safety check: PassMIT
Consult ClaudeEpicenterHQ/epicenter4.8k—~2kAutomated safety check: PassCustom licence
System Atlasinkboard/system-atlas429—~2.3kAutomated safety check: PassMIT
Design Image Studiokangarooking/design-image-studio102—~1.5kAutomated safety check: PassMIT
Kicad Reviewmixelpixx/Konnect917—~3.2kAutomated safety check: PassAGPL-3.0
Design AuditUniClipboard/UniClipboard1.9k—~554Automated safety check: PassAGPL-3.0

Similar skills

  • Consult Claude

    EpicenterHQ/epicenter

    Assign Claude Code a read-only investigation, recommendation, or finished text draft.

    4.8k GitHub stars~2k tokensUpdated yesterday
    Media & CreativeAuto-check passed
  • System Atlas

    inkboard/system-atlas

    Build and maintain an explorable, progressively-disclosed isometric "atlas" of a system's architecture — an interactive page (hover to read, click to pin, go inside for steps, moving data packets…

    429 GitHub stars~2.3k tokensUpdated 1 mo ago
    Media & CreativeAuto-check passed
  • Design Image Studio

    kangarooking/design-image-studio

    Directly generate design-oriented AI images with strong creative direction and prompt engineering.

    102 GitHub stars~1.5k tokensUpdated 5 mo ago
    Media & CreativeAuto-check passed
  • Kicad Review

    mixelpixx/Konnect

    Design review and validation workflow for KiCAD projects via MCP tools.

    917 GitHub stars~3.2k tokensUpdated 4 days ago
    Media & CreativeAuto-check passed
  • Design Audit

    UniClipboard/UniClipboard

    定期审计代码库的工程设计问题(高心智复杂度、单一真相源被破坏、catch-all 胖接口、死代码、散落魔法字面量、泄漏抽象、资源生命周期靠环形缓冲)与可优化点,范围限定为自上次审计以来的 git churn,每条发现都落到 file:line 并对照本项目自己的 VISION.md / 各级 AGENTS.md / memory…

    1.9k GitHub stars~554 tokensUpdated today
    Media & CreativeAuto-check passed
  • L1 AI Design Review

    PaperMoonuu/Design-workflow-skills

    L1 × AI 设计评审:对已完成的单页、局部 UI 设计稿进行小型迭代评审,识别影响面、状态遗漏、文案与一致性风险,并给出 P0/P1/P2 建议和验收清单。用户提供 Figma 链接、截图、前后设计稿或可评审原型,并要求设计走查、风险评审或开发前 UI 检查时使用;不用于设计前方案预检、完整多页面流程或 L2 开发交付。

    316 GitHub stars~492 tokensUpdated 17 days ago
    Media & CreativeAuto-check passed

More from magnus919/agent-skills

All 130 skills in this repo
  • Artifact Pyramids

    magnus919/agent-skills

    Organize durable agent research outputs as summaries, analysis, and evidence dossiers.

    116 GitHub stars~2.7k tokensUpdated yesterday
    Auto-check passed
  • Ascii City Engine

    magnus919/agent-skills

    Build portable, first-person colored ASCII city engines and small GIS-derived city packs.

    116 GitHub stars~1.1k tokensUpdated yesterday
    Auto-check passed
  • Color Management

    magnus919/agent-skills

    Manage color workflows with ICC profiles, working spaces, gamut mapping, and color science.

    116 GitHub stars~2.6k tokensUpdated yesterday
    Auto-check: notes
  • Data Scientist

    magnus919/agent-skills

    A skill your agent uses for PhD-level expertise in data science, statistics, and machine learning: rigorous statistical analysis, experimental design, causal inference, advanced modeling, research…

    116 GitHub stars~4.1k tokensUpdated yesterday
    Auto-check passed
  • Docker Compose

    magnus919/agent-skills

    Use Docker Compose to define, run, debug, and harden multi-container applications.

    116 GitHub stars~2k tokensUpdated yesterday
    Auto-check: notes
  • Langgraph

    magnus919/agent-skills

    Build multi-agent AI systems with LangGraph — the low-level orchestration framework for stateful, graph-based agent workflows.

    116 GitHub stars~2.8k tokensUpdated yesterday
    Auto-check passed

Questions about Fpga Development

What does Fpga Development do?

Design, review, simulate, and verify FPGA logic using explicit RTL contracts, clock and reset models, CDC analysis, timing constraints, and reproducible implementation evidence. Fpga Development is an agent skill from magnus919/agent-skills. Design, review, simulate, and verify FPGA logic using explicit RTL contracts, clock and reset models, CDC analysis, timing constraints, and reproducible implementation evidence.

When should I use Fpga Development?

Fpga Development fits situations like: synthesizable HDL; fixed-point arithmetic; resource tradeoffs; bring-up planning.

How do I install Fpga Development in Claude Code?

Run `npx skills add magnus919/agent-skills --skill fpga-development -a claude-code`. Or copy the skill folder (fpga-development in magnus919/agent-skills) into .claude/skills/fpga-development in your project. Claude Code loads it when a task matches its description.

How do I install Fpga Development in Codex?

Run `npx skills add magnus919/agent-skills --skill fpga-development -a codex`. Or copy the skill folder (fpga-development in magnus919/agent-skills) into .agents/skills/fpga-development in your project. Codex loads it when a task matches its description.

Can I use Fpga Development in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add magnus919/agent-skills --skill fpga-development -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/fpga-development, .gemini/skills/fpga-development, .github/skills/fpga-development and .opencode/skills/fpga-development in your project.

What does Fpga Development need to run?

Going by SKILL.md and its folder, Fpga Development needs Python for the scripts in its folder and the command-line tools its instructions call (python3). Our summary lists: Python 3.

Does Fpga Development access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Fpga Development safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.

What licence does Fpga Development use?

Fpga Development is published under the MIT licence (declared in SKILL.md). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Fpga Development use?

About 2.7k tokens (SKILL.md is roughly 11k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 8.8k tokens, read only when the agent opens those files.

What are the alternatives to Fpga Development?

Skills that share tags, products or a category with Fpga Development: Consult Claude (EpicenterHQ/epicenter, 4.8k stars), System Atlas (inkboard/system-atlas, 429 stars), Design Image Studio (kangarooking/design-image-studio, 102 stars) and Kicad Review (mixelpixx/Konnect, 917 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Fpga Development?

magnus919 (a GitHub user) maintains it in magnus919/agent-skills, which has 116 GitHub stars. The repository holds 130 skills in this directory. The repository was last updated on October 8, 2026.

Source: magnus919/agent-skills on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.