@namewta/speculo 2.0.6 → 2.1.0

package/skills-hub/writing/job-application/scripts/render_resume.py

@@ -0,0 +1,144 @@
+#!/usr/bin/env python3
+"""Render a tailored Markdown resume to HTML, and DOCX/PDF when libraries exist."""
+
+from __future__ import annotations
+
+import argparse
+import html
+import json
+import re
+import sys
+from pathlib import Path
+
+CLAIM_COMMENT = re.compile(r"<!--\s*claim:C-\d+\s*-->", re.IGNORECASE)
+
+
+def md_to_blocks(text: str) -> list[tuple[str, str]]:
+    blocks: list[tuple[str, str]] = []
+    paragraph: list[str] = []
+
+    def flush() -> None:
+        if paragraph:
+            blocks.append(("p", " ".join(paragraph).strip()))
+            paragraph.clear()
+
+    for raw in text.splitlines():
+        line = CLAIM_COMMENT.sub("", raw).rstrip()
+        if not line.strip() or line.strip().startswith("<!--"):
+            flush()
+            continue
+        if line.startswith("### "):
+            flush()
+            blocks.append(("h3", line[4:].strip()))
+        elif line.startswith("## "):
+            flush()
+            blocks.append(("h2", line[3:].strip()))
+        elif line.startswith("# "):
+            flush()
+            blocks.append(("h1", line[2:].strip()))
+        elif line.startswith("- "):
+            flush()
+            blocks.append(("li", line[2:].strip()))
+        else:
+            paragraph.append(line.strip())
+    flush()
+    return blocks
+
+
+def to_html(blocks: list[tuple[str, str]], css: str) -> str:
+    parts = [
+        "<!DOCTYPE html>",
+        '<html lang="zh">',
+        "<head>",
+        '<meta charset="utf-8">',
+        "<title>Resume</title>",
+        f"<style>\n{css}\n</style>",
+        "</head>",
+        "<body>",
+        '<main class="sheet">',
+    ]
+    list_open = False
+    for kind, content in blocks:
+        safe = html.escape(content)
+        if kind != "li" and list_open:
+            parts.append("</ul>")
+            list_open = False
+        if kind == "li":
+            if not list_open:
+                parts.append("<ul>")
+                list_open = True
+            parts.append(f"<li>{safe}</li>")
+        elif kind in {"h1", "h2", "h3", "p"}:
+            parts.append(f"<{kind}>{safe}</{kind}>")
+    if list_open:
+        parts.append("</ul>")
+    parts.extend(["</main>", "</body>", "</html>", ""])
+    return "\n".join(parts)
+
+
+def try_docx(blocks: list[tuple[str, str]], dest: Path) -> str:
+    try:
+        from docx import Document
+    except ImportError:
+        return "skipped: python-docx not installed"
+    doc = Document()
+    for kind, content in blocks:
+        if kind == "h1":
+            doc.add_heading(content, level=0)
+        elif kind == "h2":
+            doc.add_heading(content, level=1)
+        elif kind == "h3":
+            doc.add_heading(content, level=2)
+        elif kind == "li":
+            doc.add_paragraph(content, style="List Bullet")
+        else:
+            doc.add_paragraph(content)
+    doc.save(dest)
+    return "written"
+
+
+def try_pdf(html_path: Path, dest: Path) -> str:
+    try:
+        from weasyprint import HTML
+    except ImportError:
+        return "skipped: weasyprint not installed"
+    HTML(filename=str(html_path)).write_pdf(str(dest))
+    return "written"
+
+
+def main() -> None:
+    parser = argparse.ArgumentParser(description=__doc__)
+    parser.add_argument("--md", required=True)
+    parser.add_argument("--css", required=True)
+    parser.add_argument("--out-dir", required=True)
+    args = parser.parse_args()
+
+    md_path = Path(args.md)
+    css_path = Path(args.css)
+    out_dir = Path(args.out_dir)
+    if not md_path.is_file():
+        print(f"missing markdown: {md_path}", file=sys.stderr)
+        raise SystemExit(2)
+    css = css_path.read_text(encoding="utf-8") if css_path.is_file() else "body{font-family:sans-serif}"
+    out_dir.mkdir(parents=True, exist_ok=True)
+    blocks = md_to_blocks(md_path.read_text(encoding="utf-8"))
+    html_path = out_dir / "resume-tailored.html"
+    html_path.write_text(to_html(blocks, css), encoding="utf-8")
+    docx_status = try_docx(blocks, out_dir / "resume-tailored.docx")
+    pdf_status = try_pdf(html_path, out_dir / "resume-tailored.pdf")
+    status = {
+        "html": "written",
+        "docx": docx_status,
+        "pdf": pdf_status,
+    }
+    (out_dir / "render-status.json").write_text(
+        json.dumps(status, ensure_ascii=False, indent=2) + "\n",
+        encoding="utf-8",
+    )
+    print(json.dumps(status, ensure_ascii=False))
+    if docx_status != "written" and pdf_status != "written":
+        raise SystemExit(3)
+
+
+if __name__ == "__main__":
+    main()

package/skills-hub/writing/job-application/scripts/validate_claims.py

@@ -0,0 +1,52 @@
+#!/usr/bin/env python3
+"""Fail if tailored text cites a claim id missing from the evidence map."""
+
+from __future__ import annotations
+
+import argparse
+import re
+import sys
+from pathlib import Path
+
+CLAIM_RE = re.compile(r"claim:(C-\d+)", re.IGNORECASE)
+ID_RE = re.compile(r"\b(C-\d+)\b")
+
+
+def known_ids(evidence_text: str) -> set[str]:
+    return {match.group(1).upper() for match in ID_RE.finditer(evidence_text)}
+
+
+def main() -> None:
+    parser = argparse.ArgumentParser(description=__doc__)
+    parser.add_argument("--evidence", required=True)
+    parser.add_argument("--files", nargs="+", required=True)
+    args = parser.parse_args()
+
+    evidence = Path(args.evidence)
+    if not evidence.is_file():
+        print(f"missing evidence map: {evidence}", file=sys.stderr)
+        raise SystemExit(2)
+    allowed = known_ids(evidence.read_text(encoding="utf-8"))
+    orphans: list[str] = []
+    used = 0
+    for raw in args.files:
+        path = Path(raw)
+        if not path.is_file():
+            print(f"missing file: {path}", file=sys.stderr)
+            raise SystemExit(2)
+        text = path.read_text(encoding="utf-8")
+        for match in CLAIM_RE.finditer(text):
+            used += 1
+            claim = match.group(1).upper()
+            if claim not in allowed:
+                orphans.append(f"{path}: {claim}")
+    if orphans:
+        print("orphan claims:", file=sys.stderr)
+        for item in orphans:
+            print(f"  {item}", file=sys.stderr)
+        raise SystemExit(1)
+    print(f"ok claims={used} evidence_ids={len(allowed)}")
+
+
+if __name__ == "__main__":
+    main()

package/skills-hub/writing/job-application/scripts/write_config.py

@@ -0,0 +1,72 @@
+#!/usr/bin/env python3
+"""Write job-application path pointer JSON into the skill dir and persist_root."""
+
+from __future__ import annotations
+
+import argparse
+import json
+import sys
+from datetime import datetime, timezone
+from pathlib import Path
+
+
+def die(message: str, code: int = 2) -> None:
+    print(message, file=sys.stderr)
+    raise SystemExit(code)
+
+
+def abs_dir(raw: str, label: str, must_exist: bool) -> Path:
+    path = Path(raw).expanduser().resolve()
+    if path.exists() and not path.is_dir():
+        die(f"{label} is not a directory: {path}")
+    if must_exist and not path.is_dir():
+        die(f"{label} does not exist: {path}")
+    return path
+
+
+def inside(child: Path, parent: Path) -> bool:
+    try:
+        child.resolve().relative_to(parent.resolve())
+        return True
+    except ValueError:
+        return False
+
+
+def main() -> None:
+    parser = argparse.ArgumentParser(description=__doc__)
+    parser.add_argument("--skill-dir", required=True)
+    parser.add_argument("--persist-root", required=True)
+    parser.add_argument("--kb-root", required=True)
+    parser.add_argument("--locale", default="zh")
+    parser.add_argument("--page-budget", type=int, default=2)
+    args = parser.parse_args()
+
+    skill_dir = abs_dir(args.skill_dir, "skill-dir", must_exist=True)
+    if not (skill_dir / "SKILL.md").is_file():
+        die(f"skill-dir has no SKILL.md: {skill_dir}")
+    kb_root = abs_dir(args.kb_root, "kb-root", must_exist=True)
+    persist_root = abs_dir(args.persist_root, "persist-root", must_exist=False)
+    if inside(persist_root, skill_dir):
+        die("persist-root must not live inside the skill directory")
+    if args.page_budget < 1 or args.page_budget > 4:
+        die("page-budget must be 1..4")
+
+    persist_root.mkdir(parents=True, exist_ok=True)
+    payload = {
+        "schema_version": 1,
+        "skill": "job-application",
+        "persist_root": str(persist_root),
+        "kb_root": str(kb_root),
+        "locale": args.locale,
+        "page_budget": args.page_budget,
+        "updated_at": datetime.now(timezone.utc).strftime("%Y-%m-%dT%H:%M:%S.000Z"),
+    }
+    text = json.dumps(payload, ensure_ascii=False, indent=2) + "\n"
+    targets = [skill_dir / "config.json", persist_root / "config.json"]
+    for target in targets:
+        target.write_text(text, encoding="utf-8")
+        print(target)
+
+
+if __name__ == "__main__":
+    main()

package/skills-hub/writing/job-application/templates/application-index.md

@@ -0,0 +1,10 @@
+# 投递
+
+- company:
+- role:
+- date:
+- mode:
+- tailored:
+- speech:
+- questions:
+- improvement:

package/skills-hub/writing/job-application/templates/brief.md

@@ -0,0 +1,16 @@
+# Brief
+
+- company:
+- role:
+- locale:
+- page_budget:
+- jd_source: 用户粘贴 | 非官方原文
+- source_url:
+
+## JD 原文
+
+## 硬门槛
+
+## 核心
+
+## 加分

package/skills-hub/writing/job-application/templates/changelog.md

@@ -0,0 +1,6 @@
+# Changelog
+
+相对 `resume-original.md`，不是相对上一份别的公司的简历。
+
+| 原句 | 新句 | 理由 | claim_id | 只是重排 |
+|---|---|---|---|---|

package/skills-hub/writing/job-application/templates/company-research.md

@@ -0,0 +1,15 @@
+# 公司调研
+
+公开事实。不能写进用户的经历子弹。
+
+## 产品与岗位
+
+- 来源:
+- 日期:
+- 事实:
+
+## 与 JD 相关的技术栈
+
+## 未核实
+
+## 口播可用的一句

package/skills-hub/writing/job-application/templates/corpus.md

@@ -0,0 +1,11 @@
+# 经历原件
+
+本文件是长文原件，不是投递稿。每条可引用事实在 `evidence-map.md` 有 `C-xxx`。
+
+## 身份
+
+## 时间线
+
+## 项目与工作
+
+## 未决冲突

package/skills-hub/writing/job-application/templates/evidence-map.md

@@ -0,0 +1,5 @@
+# Evidence map
+
+| claim_id | 陈述 | source | 摘录 | 层级 |
+|---|---|---|---|---|
+| C-001 |  | src-0001 |  | 事实 |

package/skills-hub/writing/job-application/templates/framings.md

@@ -0,0 +1,6 @@
+# 预批准表述
+
+用户点头之前，敏感口径留在这里，不进投递稿。
+
+| 主题 | 允许怎么说 | 禁止怎么说 |
+|---|---|---|

package/skills-hub/writing/job-application/templates/improvement-plan.md

@@ -0,0 +1,16 @@
+# 提升计划
+
+## A. 呈现
+
+| 项 | 7 天内怎么补证 | 补到之前能否进简历 |
+|---|---|---|
+
+## B. 能力
+
+| 天数 | 要做的事 | 可验证出口 |
+|---|---|---|
+| 30 |  |  |
+| 60 |  |  |
+| 90 |  |  |
+
+## 不要做

package/skills-hub/writing/job-application/templates/ingest-log.md

@@ -0,0 +1,4 @@
+# 扫盘记录
+
+| 时间 | 文件数 | 新增 | 变化 | 备注 |
+|---|---|---|---|---|

package/skills-hub/writing/job-application/templates/known-gaps.md

@@ -0,0 +1,4 @@
+# 已知缺口
+
+| 缺口 | 为什么现在不能写进简历 | 要补的证据 |
+|---|---|---|

package/skills-hub/writing/job-application/templates/match-report.md

@@ -0,0 +1,21 @@
+# 匹配
+
+## 已覆盖
+
+| JD 要求 | claim_id | 原件位置 |
+|---|---|---|
+
+## 翻译后覆盖
+
+| JD 用词 | 语料里的真事 | claim_id |
+|---|---|---|
+
+## 诚实缺口
+
+| JD 要求 | 库里没有什么 |
+|---|---|
+
+## 不写入
+
+| 内容 | 原因 |
+|---|---|

package/skills-hub/writing/job-application/templates/methodology-cache.md

@@ -0,0 +1,10 @@
+# 方法缓存
+
+只存公开的简历与面试写法，不存用户经历。用户没有要求刷新时不要改这份文件。
+
+## 来源
+
+| 主题 | URL | 日期 | 采用的规则 |
+|---|---|---|---|
+
+## 明确不采用

package/skills-hub/writing/job-application/templates/persist-index.md

@@ -0,0 +1,11 @@
+# 岗位投递
+
+- persist_root:
+- kb_root:
+- locale:
+- page_budget:
+- corpus 更新:
+- 投递:
+
+| 日期 | 公司 | 岗位 | 目录 | 状态 |
+|---|---|---|---|---|

package/skills-hub/writing/job-application/templates/profile.md

@@ -0,0 +1,21 @@
+# Profile
+
+1. 身份
+2. 头条与目标岗位
+3. 摘要候选
+4. 经历快照
+5. 经历深挖
+6. 教育
+7. 证书与考试
+8. 技术技能
+9. 领域经验
+10. 带人与协作
+11. 公开产物
+12. 语言
+13. 奖项
+14. 志愿与社团
+15. 兴趣
+16. 到岗与工作许可
+17. 成就库
+18. 已知缺口
+19. 预批准表述

package/skills-hub/writing/job-application/templates/questions.md

@@ -0,0 +1,22 @@
+# 预测题
+
+## 开场与动机
+
+### Q1
+
+- 为什么会被问:
+- story-id:
+- 30 秒:
+- 2 分钟:
+- 追问:
+- 不会时:
+
+## 项目深挖
+
+## 技术或专业
+
+## 行为
+
+## 缺口与风险
+
+## 反问建议

package/skills-hub/writing/job-application/templates/resume-original.md

@@ -0,0 +1,10 @@
+# 原件切片
+
+本轮选用、尚未改写的经历。句子保持语料原口径，只做取舍。
+
+## 选用
+
+## 本轮不写
+
+| 内容 | 原因 |
+|---|---|

package/skills-hub/writing/job-application/templates/resume-tailored.md

@@ -0,0 +1,21 @@
+# 姓名
+
+城市 · 邮箱 · 电话 · 链接
+
+目标岗位
+
+## 摘要
+
+## 经历
+
+### 公司 · 职级
+
+时间
+
+- 做成了什么，用什么衡量，靠什么做到。 <!-- claim:C-001 -->
+
+## 项目
+
+## 技能
+
+## 教育

package/skills-hub/writing/job-application/templates/sources-index.md

@@ -0,0 +1,6 @@
+# 源索引
+
+脚本只写 `manifest.json`。这份索引由 agent 在扫盘后填写，用来决定回读哪些原文。
+
+| id | relpath | status | 要读 |
+|---|---|---|---|

package/skills-hub/writing/job-application/templates/speech.md

@@ -0,0 +1,19 @@
+# 自我介绍
+
+- 目标时长: 180–240 秒
+- 短版时长: 60–90 秒
+- 绑定投递稿:
+- 绑定故事:
+
+## 口播正本
+
+## 短版
+
+## 被打断时的接法
+
+## 数字与投递稿对照
+
+| 口播里的数字 | claim_id | 投递稿位置 |
+|---|---|---|
+
+## 不要讲

package/skills-hub/writing/job-application/templates/stories.md

@@ -0,0 +1,9 @@
+# 故事库
+
+## S-01
+
+- claim:
+- Situation:
+- Task:
+- Action:
+- Result:

package/skills-hub/writing/job-application/templates/story-bank.md

@@ -0,0 +1,11 @@
+# 本轮故事
+
+只收本轮投递稿会用到的故事。完整库在 `corpus/stories.md`。
+
+## S-01
+
+- claim_id:
+- 对应 bullet:
+- 30 秒:
+- 2 分钟:
+- 不要展开:

package/skills-hub/writing/job-application/templates/strengths.md

@@ -0,0 +1,4 @@
+# 优势
+
+| 证据 claim_id | 行为 | 能力 | 可对上的岗位信号 |
+|---|---|---|---|

package/skills-hub/writing/job-application/templates/style-memory.md

@@ -0,0 +1,7 @@
+# 风格
+
+- locale:
+- page_budget:
+- 禁用词:
+- 是否要照片:
+- 是否要英文稿:

package/skills-hub/writing/ste-zh/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 Ting'an Wang
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

package/skills-hub/writing/ste-zh/NOTICE.md

@@ -0,0 +1,9 @@
+# 来源与权利
+
+本技能是 wta-skills-hub 的独立写作技能。它不是 ASD-STE100 官方文件，也不构成合规认证。
+
+- 中文汇报规则改编自 [dualface/ste-zh](https://github.com/dualface/ste-zh)（MIT，Copyright (c) 2026 dualface）。规则编号 R1–R25 沿用该项目的自定编号，不对应 ASD-STE100 原文条号。
+- 英文模式吸收公开 skill 的工作方式：strict / STE-flavored（danyuchn/asd-ste100-skill）、80% pragmatic（JAICHANGPARK/ASD-STE100，受 Karpathy 公开建议启发）、意义优先于用词（nuelcyoung/asd-ste100）。只吸收模式，不复制这些仓库的词典或检查器。
+- 受控中文的包装动词与长定语检查点参考 mzopedia/simplified-technical-chinese 的公开方法（规范 CC BY 4.0）。本技能不收录其词表。
+- ASD-STE100 与 Simplified Technical English 的权利属于 ASD（Aerospace, Security and Defence Industries Association of Europe）。现行公开版本为 Issue 9（2025-01-15）：Part 1 为 53 条写作规则、9 节；Part 2 为约 900 个核准词。原文免费申请：https://www.asd-ste100.org/ 。本技能不收录标准原文，也不收录核准词表。
+- 通过本技能或任何 linter 不等于 STEMG 认可的合规。最终文本由人签字。