ingest-resume
Skill0 starsChanged 2 months agoMITView on GitHub
Take a user's own resume file (.tex or .docx) into the pipeline as the working master, and check it has the sections a software-engineering resume needs. Use whenever the user attaches or points to a resume file to tailor, or at the start of apply-to-job when no master resume is set yet. Preserves the user's own template and content; it polishes an existing resume, it never writes one from scratch or invents missing sections.
- Installs packages
In this repository10 other files it gives its agents.
- answer-questionsSkill
- apply-to-jobSkill
- extract-jdSkill
- find-and-applySkill
- find-jobsSkill
- humanize-textSkill
- render-resumeSkill
- resume-fit-reportSkill
- submit-applicationSkill
- tailor-resumeSkill
The file
We split this skill into the rules it gives an agent
4 rules. About 0 are common, 1 is distinctive, and 3 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 31
Section-completeness check (for software-engineering roles): Read the working master and classify each section:
- Very specific: line 40
Establish the page budget: Ask the user once how many pages their master resume is (1, 2 or 3+).
- Very specific: line 52
Write ingest-report.md to the scratchpad and report:
- Very specific: line 67
Preserve the user's template: Never restyle a .tex into a different design.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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
ingest-resume
Bring the user's resume into the pipeline as the working master, in their own template, and give an honest read on whether it has what a software-engineering resume needs. This skill does not design or author a resume: apply-kit imposes no template, and missing content is reported, never fabricated.
Input
- A path to a resume file:
.texor.docx..pdfis not supported yet (a PDF has no layout structure to preserve; see the roadmap). If given a PDF, say so and ask for the.texor.docxsource, or for the content pasted as text.
Procedure
- Detect the format by extension.
- Produce the working master in the session scratchpad as
master.tex, preserving the user's template:.tex→ copy the file verbatim. This fully preserves their template, macros, and design. Do not reformat it..docx→ convert withpandoc "<file>" -o master.tex --standalone. This preserves content and structure (headings, bold, bullets) but not the exact Word styling; tell the user the converted LaTeX is a faithful-content starting point, not a pixel copy of their Word layout. Ifpandocis not installed, stop and give the install hint (brew install pandoc).
- Section-completeness check (for software-engineering roles). Read the
working master and classify each section:
- Required: Contact (name plus at least one of email / phone / LinkedIn / GitHub), Experience (work history), Skills (technical skills), Education.
- Expected: Projects (strongly recommended for SWE, especially early-career).
- Optional (note only if present): Summary/Objective, Certifications, Open-source, Publications. Mark each Present / Thin / Missing. "Thin" means the section exists but is sparse (e.g. Skills with one line, Experience with no bullets).
- Establish the page budget. Ask the user once how many pages their master
resume is (1, 2 or 3+). apply-kit does not impose a one-page resume: the master
sets the length, and tailoring is never allowed to exceed it. Write that number
alone to a
.page-budgetfile next to the source resume (e.g.resume/.page-budget) so it is asked once and reused for every application.render-resumereads it; rendering the master itself corrects it to the real count. - Report, do not fix. Missing or thin sections are surfaced to the user as a heads-up so they can add real content to their own resume. Never invent a section or its content to fill a gap.
Output
Write ingest-report.md to the scratchpad and report:
- the working master path (
master.tex), the source format, and the page budget you recorded; - the section table (Present / Thin / Missing);
- any required section that is missing or thin, phrased as a question to the user ("Your resume has no Skills section — add one to your source resume if you have the content, or proceed as-is?").
Then hand off to resume-fit-report / tailor-resume, which read this working
master. If invoked from apply-to-job, return control to the orchestrator.
Guardrails
- Polish, never author. This skill preserves and organizes the user's real content. It does not write new bullets, add skills, or fill missing sections with invented material. Gaps are reported, not filled.
- Preserve the user's template. Never restyle a
.texinto a different design; apply-kit has no house template. - Keep resume data local. Convert and write only in the scratchpad; never send
resume content to any external tool. See
GUARDRAILS.md.
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.

