tailor-resume-to-job
Skill1 starsChanged 24 days agoMITView on GitHub
Tailor a resume to a specific job posting with ATSPrep - rank the posting's keywords, show which are missing, and rewrite the user's own lines in the posting's wording without inventing anything. Use when someone shares a job link or job description with their resume and asks to tailor, target, match or optimise it for that role.
In this repository4 other files it gives its agents.
- cover-letterSkill
- interview-prepSkill
- linkedin-profile-polishSkill
- resume-ats-checkSkill
The file
We split this skill into the rules it gives an agent
8 rules. About 0 are common, 4 are distinctive, and 4 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
Distinctive · estimate
This author's own choices. Worth a look.
4
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- could be automatic: a setting could enforce it
- Distinctive: line 16
Use whichever is available, in this order:
- Very specific: line 18
ATSPrep MCP tools: If the ATSPrep MCP server (https://atsprep.com/mcp) is connected, call its tools by name.
- Very specific: line 32
Call match_job with resume_text and job_description.
- Distinctive: line 33
Ask which missing keywords the user genuinely has.
- Very specific: line 35
If useful, call write_summary (AI) with resume_text and target_title for a tailored summary.
- Very specific: line 36
Run match_job again on the edited resume and show the change in coverage.
- Distinctive: line 41
Never add a number, employer, tool, credential or result the user didn't give you.
- Distinctive: line 42
Only add a missing keyword if the user confirms they really have that skill.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Tailor a resume to a job
How to call ATSPrep
Use whichever is available, in this order:
- ATSPrep MCP tools. If the ATSPrep MCP server (
https://atsprep.com/mcp) is connected, call its tools by name. Some clients add a prefix, e.g.mcp__atsprep__check_resume. - The bundled script (needs Node 18+ and network access):
node scripts/atsprep.mjs <tool> key=value key=@path/to/file.txtkey=@filesends a file's text, which is the safe way to pass a resume or job post. Write the user's text to a file in the current working directory first, and delete it when you're done.node scripts/atsprep.mjs listlists the tools. - Plain HTTPS (no Node): POST JSON-RPC to
https://atsprep.com/mcpwith headersContent-Type: application/jsonandAccept: application/json, text/event-stream, and the body{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"<tool>","arguments":{...}}}.
Arguments and limits for every tool are in references/tools.md. What leaves the machine is in references/privacy.md. Tell the user before sending their text to an AI tool the first time.
Steps
- Get the job post. For a link, call
fetch_job_postwithurl. If it says the site is blocked or the post wasn't found, ask the user to paste the post's text. - Get the resume as plain text (one heading, role or bullet per line).
- Call
match_jobwithresume_textandjob_description. Report coverage (overall, required, preferred) and list the missing keywords, required ones first, each with the posting sentence it came from. - Ask which missing keywords the user genuinely has. Never assume.
- Call
tailor_resume(AI) withresume_textandjob_description. It suggests rewrites of existing lines. Show each as original → suggestion, and keep only those the user approves. If its AI note says the model was skipped (busy, limit, budget), carry on with the rule-based result and suggest edits yourself under the honesty rules. - If useful, call
write_summary(AI) withresume_textandtarget_titlefor a tailored summary. - Run
match_jobagain on the edited resume and show the change in coverage. - Offer
open_in_studiowith the final text, so the user can format and download it.
Honesty rules
- Never add a number, employer, tool, credential or result the user didn't give you. If a line would be stronger with a number, ask for it.
- Only add a missing keyword if the user confirms they really have that skill.
- AI tool results come back already filtered for made-up facts. Hold your own edits to the same standard.
- A keyword the user doesn't have stays missing. Help them describe nearby experience truthfully instead.
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.

