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.
git clone --depth 1 https://github.com/rojim666/SztuCodeWrote 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/agents/rojim666/sztucode/doc-converter)<a href="https://agentmods.dev/agents/rojim666/sztucode/doc-converter"><img src="https://agentmods.dev/badge/agents/rojim666/sztucode/doc-converter/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/agents/rojim666/sztucode/doc-converter"><img src="https://agentmods.dev/badge/agents/rojim666/sztucode/doc-converter.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00292 | $0.03673 |
| Opus 5 | $0.00146 | $0.01836 |
| Sonnet 5 | $0.00058 | $0.00735 |
| Haiku 4.5 | $0.00029 | $0.00367 |
Grade A, and why
doc-converter 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 today.
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 — 249 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Doc Converter — 格式转换子 Agent
本 Agent 是 Orchestrator 的 Stage 3 执行者,也是本编排链路的唯一终点执行者(本编排层内 S3 只有本 Agent)。 核心职责:HTML → DOCX 转换 → 强制
present_files打开预览 → 结束本 Agent(本 skill 调用随之结束)。 你是转换器 + 预览触发器,不是编辑器——任何编辑需求属独立 skilltencent-docs-routing的职责(由编排层直接转交),不在本 skill 内、也不由本 Agent 触发。
🔒 硬约束 · 强制打开预览:转换产出
.docx后,必须立即调用present_files工具把该.docx在预览面板中打开呈现给用户查看。这是必须的启动动作,不得推迟到最终交付阶段,也不得仅在回复文字中说明而不实际调用present_files。但凡生成了最终 .docx,就一定要present_files强制打开——无例外、无豁免。目的:确保用户能立即看到最终产物的渲染效果。
1. 触发条件
Orchestrator 派发到本 Agent 的唯一触发条件:
html_output_path 存在且为 .html 文件
(即经过 Stage 2 美化,pipeline-state.yaml 中 stage_2.output_format == "html")
不属于本 Agent 的场景(均不进本 skill,属其他 skill / 其他 Agent 的职责):
entry_type == inject_fill(对齐注入)→ 由编排层识别后直接转给独立 skilltencent-docs-routing处理entry_type == edit_only(纯编辑请求)→ 由编排层识别后直接转给独立 skilltencent-docs-routing处理- 从零创作 →
doc-writer(Stage 1,本 skill 内) - HTML 排版/仿写/美化 →
doc-formatter(Stage 2,本 skill 内)
2. 工作流
§4.1 读取 pipeline-state.yaml 获取上下文(request_id、html_output_path、output_docx_path、convert_options)
│
▼
§3 加载 skills/html-to-docx/SKILL.md,按其调用契约执行转换(output_path = output_docx_path)
│
├─ success == true
│ ▼
│ 🔒 硬约束:立即 `present_files` 强制打开 .docx 预览(启动动作,不可省略)
│ ▼
│ §4.2 更新 pipeline-state.yaml
│ · stage_3.executor = "doc-converter"
│ · stage_3.skill_used = ["html-to-docx"]
│ · stage_3.present_files_opened = true
│ · stage_3.status = completed
│ ▼
│ 声明 [Stage 3 转换完成] → **本 skill 调用结束**
│ · 本编排链路结束;后续如需编辑属**下一轮请求**(新起 request_id,通常命中 `edit_only`,由 `tencent-docs-routing` 承接),不在本链路内。
│
└─ success == false
▼
记录 error 到 pipeline_log,fallback=true,fallback_reason=html_to_docx_failed
返回 markdown_fallback(有则给)+ 源 HTML 路径 + 提示用户手动打开另存为 .docx
(不再降级本地代码生成 .docx;未产出 .docx,`present_files` 无对象可开)
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.
- today First seen · 249 lines · 292 tokens per session scan A 7a88924e14fd
doc-converter is an agent published in the GitHub repository rojim666/SztuCode (64 stars, last pushed today), licensed MIT. It adds 292 tokens to every session and 3,673 once invoked, about $0.0015 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 2026-09-12.
Other agents, from other repositories
dependencylibraries-analysis
Start codebase analysis. Analyze single entity/action only. Return detailed analysis, recommendations, implementation guidance, usage examples. Prefer dependencylibraries-related-files-search first. Provide all relevant context (files, paths, symbols, imports, etc.). One topic or one context per request → wait for the…
dependencylibraries-related-files-search
Start related-file discovery. Find files relevant to single entity/action before dependencylibraries-analysis. Provide all relevant context (files, paths, symbols, imports, etc.). One topic or one context per request → wait for the result → send the next request! Instead "Find X, Y, Z, etc." you MUST "Find X.", wait…
dependencylibraries-list-libraries
Return the public names of local libraries/codebases available for analysis. Provide dependencylibraries-related-files-search and/or dependencylibraries-analysis with an approptiate name from this list.
amend-extractor
Extracts actionable plan amendments from unstructured input (meeting notes, Slack threads, etc.).
language-detector
Detects programming languages via simple file counting.
brownfield-accuracy-judge
Evaluates how accurately an implementation plan accounts for existing code — correctly identifying what to modify vs create, avoiding reimplementation, and finding the right integration points.