job-application
Skill0 starsChanged 36 days agoMITView on GitHub
扫描指定知识库建成经历原件,再按岗位派生投递简历(md/docx/pdf)、3–4分钟自我介绍、经历绑定的预测题与提升计划。用户说求职、改简历、投递、JD定制、自我介绍、面试题、提升计划、经历库、知识库扫盘、/job-application 时使用。不编造知识库里没有的数字、职级或技能。
13 other projects use a version of this.
Adopted by
Most recently taken up by
- Gianni03/jobtracker-skillstheir own version, first seen Nov 8, 2026
- Monsalves-Gonzalez-N/job-application-skilltheir own version, first seen Oct 20, 2026
- adriano-labs/applyarc-claude-plugintheir own version, first seen Oct 15, 2026
- SkillMedev/job-search-kittheir own version, first seen Oct 5, 2026
- YoursSarcastically/apply-for-metheir own version, first seen Aug 19, 2026
- bmohan111/claude-skillstheir own version, first seen Aug 11, 2026
Used with
Projects with it also use job-search and interview-prep.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
proficientlyjobsSkilljob-search
Search for jobs matching my resume and preferences
Also in 3 projects that use this
AnthropicSkillinterview-prep
Create structured interview plans with competency-based questions and scorecards. Trigger with "interview plan for", "interview questions for", "how should we interview", "scorecard for", or when the user is preparing to interview candidates.
Also in 3 projects that use this
Other versions13 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 13
- YoursSarcastically/apply-for-me★ 17same name
- Gianni03/gianni-skills★ 7same name
- Gianni03/jobtracker-skills★ 7same name
- pepebot-space/skills★ 2same name
- brainsforai/brains-skills★ 1same name
- mandlcho/skills★ 1same name
- adriano-labs/applyarc-claude-plugin★ 0same name
- alinzm/claude-skills★ 0same name
- bmohan111/claude-skills★ 0same name
- jvalbuena/claude-skills★ 0same name
3 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository16 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- grok-bot-team-stewardSkill
- herdrSkill
- orcaSkill
- gitea-repoSkill
- github-repo-stewardSkill
- vscode-fullstackSkill
- photo-retouchSkill
- clash-client-profileSkill
- linux-dev-disk-cleanupSkill
- optimize-codex-configSkill
- proxy-region-localeSkill
- system-health-auditSkill
- windows-dev-disk-cleanupSkill
- ste-zhSkill
The file
Files
岗位投递与面试准备
经历原件是资产,每次投递是衍生品。本文件足够开工;细则按需读 references/。不要把用户笔记、简历正文或电话邮箱写进本技能目录。
强制入口
$SKILL_DIR 是本 SKILL.md 所在文件夹(安装后常见 ~/.agents/skills/job-application、~/.claude/skills/job-application,或项目里的对应目录)。运行根是当前会话的工作区,不是技能仓,也不是用户家目录。
每次激活先解析路径,再选模式。解析顺序(高 → 低),persist_root 与 kb_root 分开取:
- 用户本轮明确给出的路径
- 已能定位的
$persist_root/config.json $SKILL_DIR/config.json- 停下来问。禁止猜测,禁止默默写到
$HOME或技能仓库根
用户说「就用当前目录」时,persist_root = <cwd>/job-application。第一次运行必须确认 kb_root 和 persist_root,写完 JSON 再扫盘。
确认后双写指针(只含路径和偏好,不含经历):
python3 "$SKILL_DIR/scripts/write_config.py" \
--skill-dir "$SKILL_DIR" \
--persist-root "<abs>" \
--kb-root "<abs>" \
--locale zh \
--page-budget 2python3 不可用时改 python。脚本把路径展开成绝对路径,写入 $SKILL_DIR/config.json 和 $persist_root/config.json。前者被技能内 .gitignore 忽略,不要提交。
模式路由
| 本轮输入 | 模式 |
| 只有知识库路径 | ingest 然后 master,停下来问岗位 |
| 已有 corpus,给了 JD / 岗位 | pack |
| 两者都有,且 manifest 里的 hash 变了 | 先 ingest + master,再 pack |
| 只要自我介绍,但本轮还没有 tailored | 先 apply 再 pitch |
| 点名只要题或只要提升计划 | questions 或 improve;没有 match-report 就先 apply |
没有 corpus 不准 apply。没有 claim_id 的数字、职级、技术名不准进投递稿或自我介绍。
产物布局
只写在 persist_root。默认是 <cwd>/job-application/。
<persist_root>/
INDEX.md
config.json
corpus/
corpus.md profile.md evidence-map.md stories.md
strengths.md known-gaps.md framings.md style-memory.md
methodology-cache.md ingest-log.md
sources/manifest.json
sources/excerpts/
applications/YYYY-MM-DD-<company>-<role>/
INDEX.md brief.md company-research.md match-report.md
resume-original.md resume-tailored.md
resume-tailored.docx resume-tailored.pdf resume-tailored.html
changelog.md speech.md questions.md story-bank.md
improvement-plan.md同日同公司同岗位已存在时追加 -01,不覆盖。知识库只读:不移动、不改写用户笔记,也不把整库复制进 persist_root。
模式
ingest
kb_root必须是已存在的目录,否则blocked-input。- 跑
scripts/ingest_kb.py --kb-root ... --out "$persist_root/corpus/sources/manifest.json"。 - 先读 manifest 和
sources/INDEX.md,再按关键词回读少量原文。单文件文本超过约 200KB 只摘头尾,摘录放sources/excerpts/。禁止把整个知识库灌进上下文。
master
从 manifest 选源,按 templates/ 写出 corpus.md、profile.md(19 节标题必须齐全)、evidence-map.md、stories.md、known-gaps.md、framings.md。成就库每条有 C-xxx。同一事实两个数字时写入缺口并问用户一次,不取较大的那个。用户明确要求时才上网刷新 methodology-cache.md;网搜结果不得变成个人成就。冲突以源文件和 evidence-map 为准,不以写得更顺的句子为准。
apply
冻结 JD 到 brief.md。先写 resume-original.md(本轮选用的原件切片),再写 resume-tailored.md。每条子弹行尾加 <!-- claim:C-012 -->。changelog.md 写原句、新句、理由、claim_id。match-report.md 四表:已覆盖、翻译后覆盖、诚实缺口、明确不写入。然后:
python3 "$SKILL_DIR/scripts/render_resume.py" \
--md "$APP/resume-tailored.md" \
--css "$SKILL_DIR/assets/resume.css" \
--out-dir "$APP"
python3 "$SKILL_DIR/scripts/validate_claims.py" \
--evidence "$persist_root/corpus/evidence-map.md" \
--files "$APP/resume-tailored.md"渲染缺依赖时 HTML/md 仍算内容完成,在回复里写降级,不要编一版「看起来像 PDF」的假文件。
pitch
读本轮 tailored、brief、story-bank,不按公司文化凭空写通用稿。speech.md 含 180–240 秒正本、60–90 秒短版、被打断时的接法、数字对照表、不要讲的内容。数字必须和投递稿同一 claim。
questions
题从四条交叉来:JD 必须项、本轮每条 bullet、被压缩但库里有的项目、诚实缺口。每题绑定 story-id,或明确写「无法作答」。缺口题只给诚实口径,不编故事。
improve
A 呈现层:本周和 7 天内能改的口径、要补的证据。B 能力层:30/60/90 天可验证出口。禁止建议伪造证书、改在职时间、把他人成果写成独立完成。
pack
顺序固定:research(写入 company-research.md)→ apply → 渲染 → pitch → questions → improve → validate_claims.py。中途失败停在当前步,不删已通过的上游文件。只有岗位名、没有 JD 原文时,可以有限检索补 JD,并在 brief 里标明「非官方原文」。
硬规则
- 指针只写两份
config.json。简历、语料、PDF 只写persist_root。 - 无
claim_id不得出口数字、职级、技术名。缺数字就定性或标needs_proof。 - 四层语言:事实 / 估计 / 解释 / 呈现选择。投递稿只用事实和已标记的估计。
- 翻译不是造假。把「写过内部工具」说成「交付过平台」需要语料里真有平台级范围。
- 背调红线不可改:学历、在职时间、公司名、正式职级、论文署名、开源仓库所有权。
- JD 有、库无 → 缺口表,不进技能栏。
- 新事实可以回写 corpus,但必须新增 source 和 claim,并在 changelog 说明。不得在 tailor 时偷偷改原件数字。
- 公司新闻不是用户的个人成就。
- 不自动调用本仓库其它 skill。不把
config.json或用户笔记提交进 git。
完整口径见 references/honesty-gate.md。简历句式、ATS、中文页数见 references/resume-playbook.md。口播结构见 references/interview-speech.md。
回复格式
每次返回:
- 模式
persist_root与kb_root(绝对路径)- 已写文件
- 跑过的脚本和退出码
- 未关闭的
needs_proof - 下一步:继续本技能的哪个模式,或结束
只有 pack 完成门全过才能说这套投递包做完了。只跑 pitch 时,完成只表示自我介绍写完。
资源
| 文件 | 何时读 |
| references/persist-paths.md | 路径、JSON 字段、第一次激活 |
| references/honesty-gate.md | 写任何对外句子之前 |
| references/ingest-protocol.md | ingest |
| references/corpus-schema.md | master / profile 19 节 |
| references/resume-playbook.md | apply |
| references/jd-parse.md | 拆 JD |
| references/interview-speech.md | pitch |
| references/question-bank.md | questions |
| references/improvement-plan.md | improve |
| references/rendering.md | 渲染降级 |
| references/company-research.md | pack 里的公司调研 |
| scripts/ | 扫盘、写指针、渲染、校验主张 |
| templates/ | 每个产物的标题骨架 |
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

