@getkrafter/resume-toolkit 1.0.3 → 1.0.5

package/README.md

@@ -6,29 +6,14 @@ Deterministic resume scoring, ATS keyword matching, and AI-powered resume tailor
 
 ## Quick Start — Skills (no account needed)
 
-### Claude Code
+### All platforms (Claude Code, Cursor, Windsurf, OpenCode, Codex)
 
 ```bash
-/install-plugin @getkrafter/resume-toolkit
+npx skills add getkrafter/resume-toolkit
 ```
 
 Then use `/score` or `/tailor` in any conversation.
 
-### Cursor
-
-Add to Settings → Remote Rule:
-
-```
-https://github.com/getkrafter/resume-toolkit/blob/master/skills/score/SKILL.md
-https://github.com/getkrafter/resume-toolkit/blob/master/skills/tailor/SKILL.md
-```
-
-Or copy the `skills/` directory to `.cursor/skills/` in your project.
-
-### Windsurf
-
-Copy the `skills/` directory to `.windsurf/skills/` or `.agents/skills/` in your project.
-
 ## Quick Start — MCP Server
 
 ### Without API key (scoring tools only)

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@getkrafter/resume-toolkit",
-  "version": "1.0.3",
+  "version": "1.0.5",
   "description": "Deterministic resume scoring, ATS keyword matching, and AI-powered resume tailoring",
   "type": "module",
   "exports": {

package/skills/score/SKILL.md

@@ -69,9 +69,11 @@ You do NOT need to do this manually. The `score_resume` MCP tool handles parsing
 
 ---
 
-## Step 5 — Call the Scoring Tool
+## Step 5 — Score the Resume
 
-**Standard path** (pasted text or file):
+Choose the method that matches your environment:
+
+**Option A — MCP tools available** (user has the MCP server configured):
 
 Call the `score_resume` MCP tool:
 ```json
@@ -81,9 +83,7 @@ Call the `score_resume` MCP tool:
 }
 ```
 
-**Krafter path** (user selected a Krafter resume):
-
-Call the `score_krafter_resume` MCP tool:
+Or for Krafter resumes, call `score_krafter_resume`:
 ```json
 {
   "id": "<the resume ID from list_resumes>",
@@ -91,6 +91,30 @@ Call the `score_krafter_resume` MCP tool:
 }
 ```
 
+**Option B — No MCP server** (skill installed without MCP):
+
+Run this inline script to score the resume. Replace the `resumeText` and `jdText` values with the actual content:
+
+```bash
+node --input-type=module -e "
+import { scoreResume, scoreATS } from '@getkrafter/resume-toolkit';
+
+const resumeText = process.env.RESUME_TEXT;
+const jdText = process.env.JD_TEXT || '';
+
+const lines = resumeText.split('\n');
+const bullets = lines.filter(l => /^\s*[-*•]|\d+[.)]/.test(l)).map(l => l.replace(/^\s*[-*•]\s*|\d+[.)]\s*/, '').trim());
+const sections = lines.filter(l => l.trim().length < 50 && l.trim().length > 0 && !(/^\s*[-*•]|\d+[.)]/.test(l)) && (l.trim() === l.trim().toUpperCase() || /^[A-Z][a-z]/.test(l.trim()))).map(l => l.trim().toLowerCase());
+
+const result = scoreResume({ rawText: resumeText, bullets, sections }, jdText || undefined);
+console.log(JSON.stringify(result, null, 2));
+" <<< ""
+```
+
+Pass `RESUME_TEXT` and `JD_TEXT` as environment variables containing the resume and job description text. Parse the JSON output and present results per Step 6.
+
+If `@getkrafter/resume-toolkit` is not installed as a dependency, install it first: `npm install @getkrafter/resume-toolkit`
+
 ---
 
 ## Step 6 — Present the Results

package/skills/tailor/SKILL.md

@@ -56,27 +56,39 @@ transformation internally via `toResumeData`.
 
 ### Step 4 -- Run Scoring Tools
 
-Run **both** scoring tools to get the full picture:
+Run **both** scoring tools to get the full picture. Choose the method that matches your environment:
 
-**For pasted/file resumes:**
+**Option A — MCP tools available** (user has the MCP server configured):
 
-1. `score_ats` with `{ resumeText, jdText }` -- returns `ATSResult` with:
-   - `score` (0-100)
-   - `matched` keywords (array)
-   - `missing` keywords (array)
-   - `details.bigramsMatched`, `details.unigramsMatched`, `details.bigramsMissing`, `details.unigramsMissing`
+For pasted/file resumes:
+1. `score_ats` with `{ resumeText, jdText }` -- returns `ATSResult` with `score`, `matched`, `missing`, and `details` breakdown.
+2. `score_resume` with `{ resumeText, jdText }` -- returns `ResumeScore` with `total`, `mode`, `breakdown`, `ats`, and `flags`.
 
-2. `score_resume` with `{ resumeText, jdText }` -- returns `ResumeScore` with:
-   - `total` (0-100)
-   - `mode` ("with-jd")
-   - `breakdown` per dimension (quantification, verbStrength, ats, bulletStructure, sectionCompleteness)
-   - `ats` (embedded ATSResult)
-   - `flags` (diagnostic messages)
+For Krafter resumes:
+1. `score_krafter_resume` with `{ id, jdText }` -- returns the full `ResumeScore` including the embedded `ATSResult`.
 
-**For Krafter resumes:**
+**Option B — No MCP server** (skill installed without MCP):
 
-1. `score_krafter_resume` with `{ id, jdText }` -- returns the full `ResumeScore` including the embedded `ATSResult`.
-2. Optionally also call `score_ats` directly if you need the standalone ATS breakdown.
+Run this inline script. Pass `RESUME_TEXT` and `JD_TEXT` as environment variables:
+
+```bash
+node --input-type=module -e "
+import { scoreResume, scoreATS } from '@getkrafter/resume-toolkit';
+
+const resumeText = process.env.RESUME_TEXT;
+const jdText = process.env.JD_TEXT;
+
+const lines = resumeText.split('\n');
+const bullets = lines.filter(l => /^\s*[-*•]|\d+[.)]/.test(l)).map(l => l.replace(/^\s*[-*•]\s*|\d+[.)]\s*/, '').trim());
+const sections = lines.filter(l => l.trim().length < 50 && l.trim().length > 0 && !(/^\s*[-*•]|\d+[.)]/.test(l)) && (l.trim() === l.trim().toUpperCase() || /^[A-Z][a-z]/.test(l.trim()))).map(l => l.trim().toLowerCase());
+
+const atsResult = scoreATS(resumeText, jdText);
+const scoreResult = scoreResume({ rawText: resumeText, bullets, sections }, jdText);
+console.log(JSON.stringify({ atsResult, scoreResult }, null, 2));
+" <<< ""
+```
+
+If `@getkrafter/resume-toolkit` is not installed, run `npm install @getkrafter/resume-toolkit` first. Parse the JSON output and continue to Step 5.
 
 ### Step 5 -- Gap Analysis