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-maintainnpx skills add VoidTechnology/voidtech-claude-plugins --skill prd-maintaingit 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 | $0.00071 | $0.03540 |
| Opus 5 | $0.00036 | $0.01770 |
| Sonnet 5 | $0.00014 | $0.00708 |
| Haiku 4.5 | $0.00007 | $0.00354 |
Grade A, and why
prd-maintain 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 — 142 lines — stays where its author put it; the contents beside it link to each section on GitHub.
prd-maintain
对既有 PRD 工作树做局部维护。心智是「外科手术改主本」，不是「重新产出一棵树」：不重走破题、模块划分确认和 PM 协同生成流程。要从原始需求生成新工作树，用 voidtech-product:prd-from-requirements。
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。
规则单源
质量红线、推断标记规范、期次权威、深度分级定义、变更记录四列约定，全部以 prd-from-requirements 为准，本技能不复制、不另立。执行前按需读取：
${CLAUDE_PLUGIN_ROOT}/skills/prd-from-requirements/SKILL.md的「质量红线」「推断标记规范」「深度分级与分期交付」「不生成汇总正文」「变更记录」「机械自检」「状态看板」「评审缺陷处置」- 涉及验收标准、状态机时，读同目录
references/对应指南；补写文档时用同目录templates/对应模板
两处表述冲突时，以 prd-from-requirements 为准并回报差异。
第 0 步：读工作树状态
不重读原始需求全量，先从工作树自身恢复上下文。决策主本（权威）：
README.md（主本规则与结构）00-global/requirement-traceability-matrix.md（头部声明与本次相关行）00-global/global-open-questions.md（OQ 状态）00-global/deepening-backlog.md（深化进度与缺口）——存在则读；它只在分期交付时必有，非分期、已全量验收级或早期版本的工作树可能没有。不存在时记录「无分期 backlog」照常继续，不要因此卡住 OQ 定案或需求变更。
00-global/status-dashboard.md 是生成物，只作健康快照用于定位，不作决策主本。
Git 处理（谨慎，不做隐式版本控制代理）：
- 不是 git 仓库：建议初始化并说明收益（变更历史、评审 diff、错改回滚），等待用户确认；用户拒绝则继续，不强制。
- 是 git 仓库：维护前记录
git status，维护后展示 diff 摘要。 git init、git commit只在用户明确要求时执行。
工况路由
按维护意图走对应工况；一次会话可组合多个工况，主本修改全部完成后统一执行一次收尾不变式。意图不明确时，先用一句话向用户确认归属，不要猜。
1. 深化模块（骨架级 → 验收级）
- 确定目标模块：用户指定优先；否则按
deepening-backlog.md的建议顺序取；无 backlog 时从各文档头部深度声明（可借助状态看板定位）找出骨架级模块，与用户确认目标后，先按templates/deepening-backlog.md补建 backlog 再开始。 - 只读该模块
prd.md、它引用的领域规格、追溯矩阵中对应区间指向的源需求区段。 - 按
templates/module-prd.md补全；过「跨文档一致性自检」并逐项记录证据，backlog 对应行标「待评审」。此时头部深度暂不改。 - 交独立核验（规则单源见
prd-from-requirements「深度分级与分期交付」）：委派@voidtech-product:product-manager按 DoD 与一致性自检逐项核验留证，记录写入 backlog「验收级核验记录」表。通过后才更新模块头部深度、第 13 节行级追溯、追溯矩阵区间入口标注和 backlog「已完成」；打回则修订后重新核验。做的人不能给自己认证。
2. 需求变更（新增需求或口径变更）
先判定工作树能力：根部存在 prd-worktree.json 且 capabilities.sourceSync 为 true 的是已迁移树，否则是 legacy 树。
已迁移树——变更从哪来决定入口，不手工比对源文件：
- 源文件出了新版本（如 Excel 更新）：一律交
voidtech-product:prd-sync（sync → propose → confirm），由它计算差异并经用户确认；确认后的变更集回到本工况，按下面第 3、4 步做影响面确认与主本合入。 - 无可比源文件的带外变更（邮件、口头、会议结论）：先登记，再走影响面确认与主本合入，不直接改主本：
python3 "${CLAUDE_PLUGIN_ROOT}/skills/prd-from-requirements/scripts/prd-sync.py" register-change <工作树> --change-id <id> --requirement <REQ|new> --text <文本>
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 · 142 lines · 71 tokens per session scan A f04d41820fa0
prd-maintain is a skill published in the GitHub repository VoidTechnology/voidtech-claude-plugins (2 stars, last pushed 28d ago), licensed Apache-2.0. It adds 71 tokens to every session and 3,540 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…