← 返回 Skills 市场

Career Spotlight Finder

Name: Career Spotlight Finder
Author: realzst

作者 RealZST · GitHub ↗ · v1.0.2 · MIT-0

cross-platform ✓ 安全检测通过

172

总下载

当前安装

版本数

在 OpenClaw 中安装

/install career-spotlight-finder

功能描述

Use when wanting to discover hidden strengths, industry buzzwords, and career narratives from past projects, articles, or code — for resumes, self-introducti...

使用说明 (SKILL.md)

Career Spotlight Finder

Discover hidden strengths and career narratives from your past projects.

Pipeline

Init → Analyze → Position → Synthesize → Write Copy → Review

Quick Reference

Step	Guide	Template
Init + Analyze	`guides/input-collection-guide.md`, `guides/project-analysis-guide.md`	`templates/project-analysis.md`
Position	`guides/domain-positioning-guide.md`	—
Synthesize	`guides/narrative-synthesis-guide.md`	`templates/aggregated-report.md`
Write Copy	`guides/copywriting-guide.md`	`templates/copywriting-variants.md`

Output

~/.career-spotlight/
├── analyses/    # per-project analyses
├── report.md    # aggregated career brand report
├── copies/      # resume-bullets, elevator-pitch, linkedin-summary, casual-intro
└── history/     # archived reports and prior copy

Step 0 — Init

Read guides/input-collection-guide.md Section 1 and follow its procedure to set up ~/.career-spotlight/.

Step 1 — Analyze

Read guides/input-collection-guide.md Sections 2-8 for the full procedure. Summary:

Collect sources from the user (local paths, URLs, or .docx files).
Validate and expand sources (auto-detect document collections in directories).
Ask the user to set project priorities (highlight or supporting).
Check existing analyses for staleness (via git hash, file mtime, or URL age).
Run new analyses per guides/project-analysis-guide.md, write to ~/.career-spotlight/analyses/.

Step 2 — Position

Read guides/domain-positioning-guide.md and follow Sections 2-4.
Recommend one expert framing with a distinctiveness thesis. Keep alternatives as wrappers.
Ask the user to confirm the framing before proceeding.

Step 3 — Synthesize

Read all analyses from ~/.career-spotlight/analyses/.
Read guides/narrative-synthesis-guide.md and follow its methodology.
Archive any existing report.md to ~/.career-spotlight/history/report-YYYY-MM-DDTHH-MM-SS.md.
Write the new report to ~/.career-spotlight/report.md using templates/aggregated-report.md.

Step 4 — Write Copy

Read ~/.career-spotlight/report.md.
Read guides/copywriting-guide.md and follow its methodology.
Archive any existing files in ~/.career-spotlight/copies/ to history/ with timestamp suffix.
Write four files to ~/.career-spotlight/copies/ using templates/copywriting-variants.md:
- resume-bullets.md
- elevator-pitch.md
- linkedin-summary.md
- casual-intro.md

Step 5 — Review

Present a summary: positioning statement, theme line count, top 3 hidden capabilities.

Then offer:

Add more projects → Step 1
Change domain direction → Step 2
Adjust narrative emphasis → Step 3
Regenerate copy variants → Step 4
Accept and finish — remind the user their files are at ~/.career-spotlight/

安全使用建议

What to consider before installing/running: 1) This skill will scan and read the project folders and documents you point it at and will create and write files under ~/.career-spotlight/ (analyses, report.md, copies/, history/). Do not point it at folders containing secrets, private keys, or sensitive data you don't want processed. 2) The registry metadata lists pandoc as a required binary but the README says pandoc is optional (fallback to python-docx). Install pandoc if you want robust .docx → markdown conversion, or ensure python-docx is available if you rely on the fallback. 3) The skill may fetch only the URLs you provide (WebFetch); avoid providing URLs that return private content you don't want downloaded. 4) Review the generated files after a run before sharing them externally. 5) If you are especially cautious, run the skill in a limited environment (container or VM) and grant it access only to the specific directories you want analyzed. 6) The skill claims "Everything stays local" and the instructions support that, but since this is instruction-only content from an external source, review the version on the linked repository (https://github.com/RealZST/career-spotlight-finder) if you need higher assurance that no external upload behavior has been added in later changes.

功能分析

Type: OpenClaw Skill Name: career-spotlight-finder Version: 1.0.2 The career-spotlight-finder skill bundle is a comprehensive tool designed to analyze local code repositories, academic papers, and URLs to generate professional resumes and LinkedIn profiles. While the skill requires high-privilege tools such as Bash, WebFetch, and Write, its behavior is strictly aligned with its stated purpose, and it includes explicit security instructions in 'guides/input-collection-guide.md' and 'guides/project-analysis-guide.md' to treat external data as untrusted and avoid indirect prompt injection. No evidence of data exfiltration, malicious execution, or obfuscation was found; all outputs are stored locally in '~/.career-spotlight/'.

能力评估

ℹ Purpose & Capability

The skill's name, description, and runtime instructions consistently describe analyzing local projects, documents, and URLs to produce a local report and copy; the requested capabilities (read, glob, grep, bash, write, WebFetch) fit that purpose. One mismatch: SKILL.md / registry metadata lists pandoc as an 'anyBins' requirement (effectively required), but README and the guides state pandoc is optional with a fallback to python-docx. That inconsistency could cause false expectations about availability of conversion tools.

ℹ Instruction Scope

The SKILL.md and included guides explicitly instruct the agent to read user-provided local paths, repo files, converted .docx content, saved analyses in ~/.career-spotlight/, and to fetch only user-supplied URLs. This scope matches the stated purpose, but it does involve broad local file reads (project trees, config files, and metadata such as git hashes and mtimes) and will write files under ~/.career-spotlight/. The guides include explicit skip lists (e.g., .env, node_modules, .git) and caution about treating fetched web content as untrusted, which mitigates some concerns.

✓ Install Mechanism

This is instruction-only with no install spec and no code files to execute from a downloaded archive. That is the lowest-risk install model.

✓ Credentials

The skill declares no required environment variables or credentials. It optionally lists pandoc as an external binary. There are no requests for unrelated cloud credentials or sensitive tokens.

✓ Persistence & Privilege

The skill does write and maintain a local workspace (~/.career-spotlight/) and archives prior reports, which is consistent with its function. It is not force-enabled (always: false) and does not request elevated or cross-skill configuration changes.

如何使用

确保已安装 OpenClaw（本地或 Docker 部署）
在对话框中输入安装命令：/install career-spotlight-finder
安装完成后，直接呼叫该 Skill 的名称或使用 /career-spotlight-finder 触发
根据 Skill 的参数说明提供必要输入，即可获得结构化输出

版本历史

v1.0.2

- Added detailed user and implementation guide: guides/input-collection-guide.md. - All steps for input collection and analysis now reference this new guide for consistent procedures. - Changelog focuses on improved onboarding and source validation through structured documentation.

v1.0.1

No changes detected from previous version. - Version bump only; no file changes in this release.

v1.0.0

Career Spotlight Finder v1.0.0 — Initial Release - Launches a skill for uncovering hidden strengths, industry buzzwords, and personal branding narratives from past projects, articles, or code. - Guides users through initializing a local career profile directory, collecting and validating diverse project sources (local paths, URLs, docx files), and analyzing them for career value. - Supports priority tagging of projects (highlight/supporting) to tailor narrative emphasis. - Handles directories intelligently: distinguishes between single projects and document collections. - Checks for existing or out-of-date analyses, ensuring results reflect up-to-date material. - Stores all analyses and outputs locally for privacy and easy reuse in resumes and self-introductions.

元数据

Slug career-spotlight-finder

版本 1.0.2

许可证 MIT-0

累计安装 0

当前安装数 0

历史版本数 3

常见问题