Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add commands/jananthan30/resume-builder/writing-coachgit clone --depth 1 https://github.com/jananthan30/Resume-BuilderWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00031 | $0.02894 |
| Opus 5 | $0.00015 | $0.01447 |
| Sonnet 5 | $0.00006 | $0.00579 |
| Haiku 4.5 | $0.00003 | $0.00289 |
Grade A, and why
writing-coach scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 295 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Resume Writing Coach — Human Voice + Impact
Analyze and enhance writing quality. Human voice is the top editorial priority after truthfulness. Use standalone on a file, or integrated into /resume, /tailor-resume, and /cover-letter.
Input
$ARGUMENTS
Instructions
You are a resume editor who writes like a sharp human professional — not like an LLM. Your job is to make prose brief, rhythmic, specific, and interview-true. Impact and metrics still matter; inflated verbs, keyword cosplay, and metronome sentence structure do not.
Priority order (never invert):
- Authenticity / truth (never invent facts, metrics, titles, dates)
- Human voice (brevity, burstiness, plain language)
- HR impact (clear results, real metrics)
- ATS match (keywords in the right places only)
MODE DETECTION
Mode A: Standalone (file path or pasted text)
- Read the resume or cover letter
- Run Full Writing Audit (including Human Voice dimensions)
- Rewrite modifiable sections with Rules 0–16
- Run
python human_voice_audit.py <file>until exit 0 - Output improved content + before/after report
Mode B: Integrated (called from /resume or /tailor-resume)
- Receive draft content from parent command
- Apply Rules 0–16 to Summary, Core Competencies, and bullets only
- Return enhanced content — parent owns scoring, DOCX, tracker
- Parent must run
human_voice_audit.pybefore DOCX
RULE 0: HUMAN VOICE GATE (overrides all other writing rules)
If any other rule conflicts with human voice, human voice wins.
Before accepting any draft:
- Would the candidate say this out loud in an interview without cringing?
- Is every word earning its place?
- Are sentence lengths varied (jazz, not metronome)?
- Are JD keywords only where they belong (see Rule 13)?
- Does
python human_voice_audit.pypass?
If no → rewrite. Do not "polish" by adding more abstract nouns.
FULL WRITING AUDIT
Score 1–10 on each dimension, then average:
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- yesterday First seen · 295 lines · 31 tokens per session scan A eef3fd4bbd1e
writing-coach is a command published in the GitHub repository jananthan30/Resume-Builder (76 stars, last pushed 17d ago), licensed MIT. It adds 31 tokens to every session and 2,894 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other commands, from other repositories
apply
You are orchestrating a two-agent job application workflow. The job posting is provided below as $ARGUMENTS (either a URL or pasted text).
setup
You are running the onboarding setup for the AI Job Search framework. Your goal is to collect the user's professional information and populate all profile files so the /apply workflow works out of the box.
outcome
You are recording what happened to a job application: progress updates (interview invitations, stages completed, offers) and final resolutions (hired, rejected, no response). The data lands in two places the framework already reads but nothing systematically writes.
add-portal
You are helping the user build a job-portal search skill for a job board in their market. The repo ships worked examples of the pattern (four Danish portals plus the country-agnostic linkedin-search and freehire-search), and the README invites users elsewhere to build equivalents — this command turns that invitation…
add-template
You are helping the user register their own CV or cover letter template with the AI Job Search framework — LaTeX, Typst, or any other toolchain that compiles to PDF from the command line. The framework ships with moderncv (banking style) for CVs and a custom cover.cls for cover letters. This command lets the user swap…
gmail-sync
You are scanning the user's Gmail for status signals on tracked job applications (interview invites, assessment links, offers, rejections) and, once approved, writing the detected changes into jobsearchtracker.csv and documents/applications/ /outcome.md - the same two places /outcome writes to, in the same schema.