Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add timzaak/web-dev-skills/plugin install t-toolsWrote 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/timzaak/web-dev-skills/t-dream)<a href="https://agentmods.dev/skills/timzaak/web-dev-skills/t-dream"><img src="https://agentmods.dev/badge/skills/timzaak/web-dev-skills/t-dream.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 | $0.00086 | $0.04304 |
| Opus 5 | $0.00043 | $0.02152 |
| Sonnet 5 | $0.00017 | $0.00861 |
| Haiku 4.5 | $0.00009 | $0.00430 |
Grade A, and why
t-dream scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 211 lines — stays where its author put it; the contents beside it link to each section on GitHub.
上下文整理与工程事实重组
运行时边界：${CLAUDE_PLUGIN_ROOT}/protocols/runtime-boundaries.md（写报告或调用目标项目脚本前读）
需求来源边界：${CLAUDE_PLUGIN_ROOT}/protocols/requirement-source-contract.md（区分草稿与已发布需求来源时读）
候选问题结构、评分权重、报告结构和模式写入边界：${CLAUDE_PLUGIN_ROOT}/protocols/dream-report-contract.md（构造 subagent 输出、计算评分或写报告前读）
目标
- 把 PRD、用户故事、设计、任务、Demo 注释、实现事实和项目结构重新收敛成当前可信上下文。
- 识别并减少错误信息累积：重复 PRD、历史过程文档、过期术语、失效设计、冲突业务规则、错误能力承诺和误导性实现说明。
- 评估 PRD 目录、代码目录、模块边界、测试布局、Demo 布局、
.ai/运行时产物组织是否合理，是否支持后续 AI agent 快速定位、窄范围修改和可靠验证。 - 建立 PRD -> 用户故事 -> 设计/任务 -> 代码 -> 测试/Demo 的 traceability 视图，指出断链、错链、重复链和缺失证据。
- 对比 PRD 与实现事实，检查能力边界、模型约束、校验、权限、业务逻辑、前端可见行为和 Demo 覆盖事实的差异。
- 通过专用 subagent 和多个
general_agent按维度并行检查，主流程只负责编排、验证、合并、裁决和必要写入。 - 输出可追溯证据、差异分级和应修正文档还是实现的建议。
t-dream 聚焦目标项目当前上下文是否干净、可信、结构合理，并能继续支撑后续 AI 编程流程。
参数
| 参数 | 说明 |
|---|---|
[feature] |
可选。聚焦指定功能、模块或领域关键词 |
--all |
检查全部可识别 PRD / 模块 |
--deep |
启用后端模块级深度一致性检查 |
--backend-only |
只执行 PRD 与后端实现一致性检查（隐含 --deep） |
--govern-prd |
显式进入 PRD 治理写入模式 |
默认模式是只读 audit，不修改目标项目文档或代码。未传入 [feature] 或 --all 时，先提示可用模块来源是 docs/prd/**/*.md，再建议用户指定范围。
用户只要求"检查/评估/排查准确性"时，保持只读 audit 模式；删除是高风险操作，见"PRD 治理模式"。
输入范围
- PRD：
docs/prd/**/*.md（排除模板、索引和说明文件） - 用户故事：
docs/user-stories/**/*.md、.ai/user-stories/**/*.md - PRD 草稿：
.ai/prd/**/*.md - 设计与任务：
.ai/design/**/*.md、.ai/task/** - Demo 测试：Web 为
demo/e2e/**/*.e2e.ts，Flutter 为patrol_test/**/*_test.dart - 实现代码：按目标项目真实结构定位 backend、frontend、demo 相关实现
- 项目结构：
docs/、.ai/、backend/frontend/demo/test 目录、README、AGENTS/CLAUDE 类上下文入口、ADR 或架构说明（如存在）
执行模型
t-dream 是主控编排 skill，不应由主线程独自完成所有检查。主线程负责确定范围、构造共享上下文、并行启动 subagent、验证候选问题、合并结果、写入最终报告，并在 PRD 治理模式下执行经过验证的文档改动。
整体采用类似 code review 的两阶段机制：
- 并行发现：专用 subagent 与多个
general_agent从不同维度独立发现候选问题。 - 统一验证：主线程或专门的验证 subagent 根据真实文件证据过滤误报、去重、定级和评分。
模式语义（audit / govern-prd / backend-only / deep 的写入边界和覆盖范围）以 ${CLAUDE_PLUGIN_ROOT}/protocols/dream-report-contract.md 的模式表为准。补充约定：
- 默认聚焦 context health：当前上下文是否准确、收敛、可导航、可追踪、结构上可持续。措辞、格式、风格偏好仅在导致业务含义失真、查找成本显著增加或后续 agent 容易误用时进入问题清单。
govern-prd先完成 audit 和治理计划，再按计划改写 PRD、索引和引用。backend-only适合实现阶段收口，不输出上下文治理或结构组织结论。
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.
- 2d ago Changed · -195 lines c38d26712114
- 8d ago First seen · 406 lines · 86 tokens per session scan A b281ec4c1ed0
t-dream is a skill published in the GitHub repository timzaak/web-dev-skills (71 stars, last pushed today), licensed Apache-2.0. It adds 86 tokens to every session and 4,304 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 2026-08-30.
Other skills, from other repositories
chinese-git-workflow
A reference for configuring Git with Chinese code-hosting services such as Gitee, Coding.net, GitLab China, and CNB, including SSH, HTTPS, credentials, CI, and repository mirroring.
chinese-documentation
A Chinese technical-documentation style guide covering spacing, punctuation, numbers, terminology, and links when Chinese and English appear together.
chinese-commit-conventions
A Chinese-language guide to Conventional Commits, a format for writing consistent Git commit messages, plus related changelog, commit-checking, and commit-helper configuration.
systematic-debugging
A step-by-step method for finding the underlying cause of technical problems before changing code. It covers reading errors, reproducing failures, checking recent changes, and tracing data across system components.
workflow-runner
A procedure for running agency-orchestrator YAML workflows directly inside Claude Code, OpenClaw, or Cursor. The current AI session performs each named role, so no separate API key is needed.
build-executor
Govern implementation from an approved execution contract. Invoke when execution-contract.md is approved and the user wants disciplined build work, TDD execution, or guarded batch-by-batch implementation.