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/voidtechnology/voidtech-claude-plugins/prd-from-requirementsnpx skills add VoidTechnology/voidtech-claude-plugins --skill prd-from-requirementsgit clone --depth 1 https://github.com/VoidTechnology/voidtech-claude-pluginsWhat 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 | $0.00089 | $0.10707 |
| Opus 5 | $0.00044 | $0.05353 |
| Sonnet 5 | $0.00018 | $0.02141 |
| Haiku 4.5 | $0.00009 | $0.01071 |
Grade A, and why
prd-from-requirements 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 — 498 lines — stays where its author put it; the contents beside it link to each section on GitHub.
prd-from-requirements
本技能用于从原始需求生成高质量 PRD 工作树。
执行任何随插件分发的脚本前，先读 HOST-RUNTIME.md（插件内路径：skills/_shared/HOST-RUNTIME.md），按当前宿主选择唯一入口。下文保留的 ${CLAUDE_PLUGIN_ROOT} 命令是 Claude Code 形式；OMP 必须映射为 voidtech_product_runtime 调用，不得把未展开变量交给 shell。
与 to-prd 的分工
本技能适用于“从原始需求文档生成模块化 PRD 工作树”，主要服务专业产品经理、业务分析、研发评审和测试验收。
如果用户只是希望把当前对话整理成一份轻量单体 PRD，并发布到项目 issue 追踪器，应该改用 voidtech-engineering:to-prd。to-prd 不重新访谈需求，也不产出模块化工作树。
对已生成工作树的后续维护——深化模块到验收级、合入需求变更、定案开放问题、落实评审修订——改用 voidtech-product:prd-maintain，不要重跑本技能。prd-maintain 复用本技能的红线、模板与脚本，只封装维护流程。源需求文件（如 Excel）后续出新版本时的同步走 voidtech-product:prd-sync：它只计算差异并提出候选，确认后的合入仍由 prd-maintain 执行。本技能只负责首建。
触发场景
当用户要求基于原始需求、Excel 整理稿、访谈纪要、需求清单或已有粗稿生成、重构、评审 PRD 时，使用本技能。
典型请求包括：
- 根据原始需求生成 PRD
- 把需求整理成产品需求文档
- 生成模块化 PRD 工作树
- 生成需求追溯矩阵
- 检查 PRD 是否能支撑研发和测试
- 将业务需求拆成可评审、可开发、可验收的文档
目标
把原始需求转化为高质量 PRD 工作树，而不是简单改写原文。
产出必须做到：
- 需求可追溯
- 模块边界清晰
- 跨系统依赖显式
- 核心流程可执行
- 状态机可判断
- 字段规则可落地
- 权限矩阵可测试
- 验收标准可判真伪
- 开放问题不被私自定案
输入
用户至少需要提供：
- 原始需求路径或原始需求正文
- 输出目录（可选；未提供时按“输出目录安全策略”生成默认目录）
如果需求来源包含多个文件，优先完整读取主需求文件，再读取直接引用的附件、表格、旧版 PRD 或补充说明。
Excel 输入处理
如果原始需求是 .xlsx，不要直接用 Read 读取二进制文件。目标是先得到可审阅的文本中间产物，再读取 Markdown 或 CSV。按下面的降级阶梯处理：
-
优先使用技能自带脚本。该脚本只使用 Python 标准库，不依赖 pandas、openpyxl 等第三方包。
python3 "${CLAUDE_PLUGIN_ROOT}/skills/prd-from-requirements/scripts/xlsx-to-markdown.py" \ <原始需求.xlsx> \ <输出目录>/_source/<需求文件名>/ -
如果没有
python3，探测本机常见转换工具。优先保留多 sheet 信息。command -v xlsx2csv command -v in2csv- 有
xlsx2csv时，使用xlsx2csv -a <原始需求.xlsx> <输出目录>/_source/<需求文件名>/导出所有 sheet。 - 有
in2csv时，先用in2csv -n <原始需求.xlsx>查看 sheet，再按 sheet 导出到_source/<需求文件名>/。
- 有
-
如果以上工具都不可用，停下请求用户转换。明确告知用户：请在 Excel、WPS 或 Numbers 中把需求表另存为
.csv后重新提供；拿到 CSV 前不要假装已经读取.xlsx内容。
转换产物固定放在当前 PRD 输出目录的 _source/<需求文件名>/ 下，并标注为“转换产物，以原始 xlsx 为准”：
INDEX.md：sheet 清单和转换结果索引。NN-{sheet}.md：每个 sheet 的 Markdown 表格，优先供模型阅读。NN-{sheet}.csv：每个 sheet 的 CSV，供精确检索、计数或复核。
What ships with it
60 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.
- assets/logic-atlas-viewer.html 217 KB
- assets/renderer-validation-proof.json 519 B
- references/acceptance-criteria-guide.md 1.9 KB
- references/changelog-and-review-discipline.md 2.7 KB
- references/pm-agent-collaboration.md 4.0 KB
- references/prd-quality-checklist.md 5.2 KB
- references/state-machine-guide.md 1.4 KB
- schemas/journal-record.schema.json 5.0 KB
- schemas/logic-model.schema.json 9.5 KB
- schemas/normalization-manifest.schema.json 1.3 KB
- schemas/operation.schema.json 4.1 KB
- schemas/prd-worktree.schema.json 1.5 KB
- schemas/proposal.schema.json 3.4 KB
- schemas/revision-manifest.schema.json 945 B
- schemas/source-registry.schema.json 916 B
- schemas/sync-state.schema.json 1.3 KB
- scripts/check-prd-tree.py 4.2 KB runs code
- scripts/generate-dashboard.py 33 KB runs code
- scripts/prd-sync.py 26 KB runs code
- scripts/prdsync/__init__.py 144 B runs code
- scripts/prdsync/atlas.py 176 KB runs code
- scripts/prdsync/base_cas.py 5.2 KB runs code
- scripts/prdsync/canonical_store.py 2.3 KB runs code
- scripts/prdsync/core_archify.py 3.1 KB runs code
- scripts/prdsync/effective_view.py 2.5 KB runs code
- scripts/prdsync/journal_projector.py 4.7 KB runs code
- scripts/prdsync/manifest_checks.py 1.1 KB runs code
- scripts/prdsync/markdown_validator.py 32 KB runs code
- scripts/prdsync/merge.py 30 KB runs code
- scripts/prdsync/migration.py 16 KB runs code
- scripts/prdsync/operation_engine.py 21 KB runs code
- scripts/prdsync/schema_validator.py 4.6 KB runs code
- scripts/prdsync/sync.py 26 KB runs code
- scripts/prdsync/writer_lock.py 6.6 KB runs code
- scripts/xlsx-to-markdown.py 7.9 KB runs code
- templates/cross-system-dependencies.md 3.1 KB
- templates/cross-system-flows.md 3.3 KB
- templates/deepening-backlog.md 4.3 KB
- templates/domain-spec.md 2.7 KB
- templates/feature-gating-matrix.md 2.0 KB
- templates/global-open-questions.md 2.0 KB
- templates/glossary.md 2.2 KB
- templates/module-prd.md 17 KB
- templates/prd-worktree/README.md 2.5 KB
- templates/product-overview.md 3.2 KB
- templates/requirement-traceability-matrix.md 2.9 KB
- templates/system-overview.md 1.7 KB
- tests/legacy_fixture.py 13 KB runs code
- tests/test_archify_bridge.py 17 KB runs code
- tests/test_archify_css.py 5.5 KB runs code
- tests/test_architecture_ir.py 15 KB runs code
- tests/test_check_prd_tree.py 26 KB runs code
- tests/test_cli.py 9.4 KB runs code
- tests/test_dashboard_guards.py 3.9 KB runs code
- tests/test_effective_view.py 2.3 KB runs code
- tests/test_gate2_example.py 2.7 KB runs code
- tests/test_gate2_migration.py 5.9 KB runs code
- tests/test_gate3_rebaseline.py 4.7 KB runs code
- tests/test_gate3_sync.py 8.0 KB runs code
- tests/test_gate4_merge.py 10 KB runs code
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 · 498 lines · 89 tokens per session scan A c5b45dbe59b8
prd-from-requirements is a skill published in the GitHub repository VoidTechnology/voidtech-claude-plugins (2 stars, last pushed 28d ago), licensed Apache-2.0. It adds 89 tokens to every session and 10,707 once invoked, about $0.0004 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
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
babysit-pr
Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…
imagegen
Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…