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.
git clone --depth 1 https://github.com/MadeByTokens/resume-helperWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/madebytokens/resume-helper/help)<a href="https://agentmods.dev/commands/madebytokens/resume-helper/help"><img src="https://agentmods.dev/badge/commands/madebytokens/resume-helper/help/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/madebytokens/resume-helper/help"><img src="https://agentmods.dev/badge/commands/madebytokens/resume-helper/help.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What 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 5.1 | $0.00007 | $0.01159 |
| Opus 5 | $0.00003 | $0.00580 |
| Sonnet 5 | $0.00001 | $0.00232 |
| Haiku 4.5 | $0.00001 | $0.00116 |
Grade A, and why
help 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 9d ago.
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 — 138 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Resume Helper - Help
Display the following help information:
╔══════════════════════════════════════════════════════════════════╗
║ RESUME HELPER PLUGIN ║
║ Adversarial multi-agent resume development for Claude Code ║
╚══════════════════════════════════════════════════════════════════╝
COMMANDS
────────────────────────────────────────────────────────────────────
/resume-helper:resume-loop <experience_file> [options]
Start the adversarial resume development loop.
Arguments:
<experience_file> Path to your experience/background file (required)
Options:
--job <file> Path to job description for targeting
--max-iterations <n> Maximum improvement cycles (default: 5)
--max-pages <n> Maximum pages: 1, 2, or 3 (default: 1, prompts for confirmation)
--output <file> Output path for final resume (default: ./resume_final.md)
--premium Use Opus model for Coach (higher quality, higher cost)
Examples:
/resume-helper:resume-loop "experience.md"
/resume-helper:resume-loop "exp.md" --job "job.md"
/resume-helper:resume-loop "exp.md" --job "jd.md" --max-pages 2
/resume-helper:resume-loop "exp.md" --job "jd.md" --premium
────────────────────────────────────────────────────────────────────
/resume-helper:resume-status
Check status of active or completed loop.
Shows current phase, iteration, and outstanding concerns.
────────────────────────────────────────────────────────────────────
/resume-helper:cancel-resume
Cancel an active loop. Progress is preserved in working/state.json.
────────────────────────────────────────────────────────────────────
/resume-helper:interview-prep [resume_file]
Generate interview preparation document.
Arguments:
[resume_file] Path to resume (optional - uses last loop result if omitted)
Output:
working/output/interview_prep.md - Contains likely questions and suggested responses
────────────────────────────────────────────────────────────────────
QUICK START
────────────────────────────────────────────────────────────────────
1. Create experience.md with your raw experience
2. Run: /resume-helper:resume-loop "experience.md"
3. Answer the Coach's clarifying questions
4. Get your output files:
- resume_final.md (your polished resume)
- working/output/interview_prep.md (preparation guide)
- working/coach/assessment.md (full audit trail)
- working/inputs/candidate_additions.md (your answers to Coach questions)
NOTE: Default is 1-page resume. Use --max-pages 2 or --max-pages 3 for longer resumes.
Recommended: 1 page (<10 yrs exp), 2 pages (10-20 yrs), 3 pages (executives/academics)
────────────────────────────────────────────────────────────────────
HOW IT WORKS
────────────────────────────────────────────────────────────────────
The loop uses 4 agents with different roles:
Writer → Creates/improves the resume (advocates for you)
Fact-Checker → Verifies claims against your input (catches hallucinations)
Interviewer → Reviews like a hiring manager (finds weaknesses)
Coach → Synthesizes feedback and asks clarifying questions
Flow per iteration:
1. Writer creates resume draft
2. Fact-Checker verifies all claims (blocks hallucinations)
3. Interviewer reviews skeptically
4. Analysis agents check for issues (parallel)
5. Coach synthesizes and provides guidance
────────────────────────────────────────────────────────────────────
ARCHITECTURE: FILE-BASED MESSAGE PASSING
────────────────────────────────────────────────────────────────────
All agents communicate through files to minimize context window usage:
working/
├── inputs/ # Your input files
│ ├── experience.md # Original experience
│ ├── job_description.md # JD (if provided)
│ └── candidate_additions.md # Your answers to questions
├── writer/ # Writer's outputs
│ ├── output.md # Current resume draft
│ ├── notes.md # Writer's notes about changes
│ └── status.md # Status: DONE or BLOCKED
├── fact_checker/ # Fact-Checker's outputs
│ ├── report.md # Full verification report
│ └── verdict.md # PASS or FAIL
├── interviewer/ # Interviewer's outputs
│ ├── review.md # Full review
│ └── verdict.md # STRONG_CANDIDATE/NEEDS_WORK/RED_FLAGS
├── analysis/ # Analysis outputs
│ ├── vague_claims.md
│ ├── buzzwords.md
│ ├── ats_compatibility.md
│ └── quantification.md
├── coach/ # Coach's outputs
│ ├── assessment.md # Full assessment
│ ├── feedback.md # Feedback for Writer
│ ├── questions.md # Questions for you
│ └── verdict.md # READY/NEEDS_STRENGTHENING/etc
└── state.json # Loop state
This architecture prevents context exhaustion in long loops.
────────────────────────────────────────────────────────────────────
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.
- 9d ago First seen · 138 lines · 7 tokens per session scan A a0e1240b8872
help is a command published in the GitHub repository MadeByTokens/resume-helper (7 stars, last pushed 7mo ago), licensed MIT. It adds 7 tokens to every session and 1,159 once invoked, about $0.0000 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 2026-08-31.
Other commands, from other repositories
learn
Force claude-smart to extract learnings from this session now.
prototype
You are building a proof-of-concept for the current Grainulator sprint. Read CLAUDE.md for sprint context and claims.json for existing research claims.
init
Install the formatters this repository needs, with every command visible before it runs.
brand-generate
Generate an on-brand document from a saved Brand Profile.
genshijin-compress
A command for safely shortening Markdown or text files into the genshijin style.
adr-supersede
Create a new ADR that supersedes an existing one.