Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/leopu00/job-hunter-teamnpx agentmods add skills/leopu00/job-hunter-team/scaling-calcWrote 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/skills/leopu00/job-hunter-team/scaling-calc)<a href="https://agentmods.dev/skills/leopu00/job-hunter-team/scaling-calc"><img src="https://agentmods.dev/badge/skills/leopu00/job-hunter-team/scaling-calc/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/skills/leopu00/job-hunter-team/scaling-calc"><img src="https://agentmods.dev/badge/skills/leopu00/job-hunter-team/scaling-calc.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00047 | $0.00909 |
| Opus 5 | $0.00023 | $0.00454 |
| Sonnet 5 | $0.00009 | $0.00182 |
| Haiku 4.5 | $0.00005 | $0.00091 |
Grade A, and why
scaling-calc 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 12d 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 — 67 lines — stays where its author put it; the contents beside it link to each section on GitHub.
🎚️ scaling-calc — change gear one step at a time, not straight into 6th
When the team opens the work window (or you need to consume more), do NOT set off in 6th gear ("plenty of budget → spawn 5 scouts / throttle to 0"): you do not yet know how much a worker really consumes in THIS cycle. You calibrate by steps.
Procedure
1. Start with 1 SINGLE worker at the floor (5min, the minimum for workers).
2. Observe for ~30 min to measure the real burn. Read the worker's burn:
python3 /app/shared/skills/rate_budget.py # sustainable target speed (S)
# per-agent burn: from the table the Sentinella forwards you, or:
python3 /app/shared/skills/agent-speed-table.py
Take: S = sustainable speed (e.g. sustainable_burn %weekly/h) and b = the
worker's measured burn (same unit).
3. Compute roster + throttle:
python3 /app/agents/_skills/scaling-calc/scaling_calc.py --target <S> --measured <b>
# if you observed N workers at throttle T:
python3 .../scaling_calc.py --target <S> --measured <b_total> --workers <N> --throttle <T>
It gives you: how many workers, which throttle, and a staged plan.
4. Spawn IN STAGES following the plan: one at a time, re-measuring before the next one (~10 min is enough to see the newcomer's burn). NEVER spawn the whole block in one go.
Those 10 minutes are an observation window, not a phase offset: the phase distance between two workers on the same step is
T/N(the period divided by the number of workers sharing it) and the launcher applies it by itself at spawn time. It is not a number to decide here, and it is not a constant: on a 5-minute step, three workers want to be 100s apart from each other.
The two levers
- Worker under target (1 worker burns less than the target) → the lever is the number of workers (parallelism), all at the floor. Add them in stages.
- Worker over target (1 worker already burns more than the target) → the lever is the throttle: keep 1 worker and raise its throttle (the tool gives you the exact value). NEVER zero the throttle out (workers have a 5min floor anyway).
What ships with it
7 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 12d ago First seen · 67 lines · 47 tokens per session scan A a5a1040a2b74
scaling-calc is a skill published in the GitHub repository leopu00/job-hunter-team (49 stars, last pushed yesterday), licensed MIT. It adds 47 tokens to every session and 909 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 skills, from other repositories
persona-event-coordinator
Plan and manage events — scheduling, invitations, and logistics.
persona-researcher
Organize research — manage references, notes, and collaboration.
persona-team-lead
Lead a team — run standups, coordinate tasks, and communicate.
recipe-review-meet-participants
Review who attended a Google Meet conference and for how long.
recipe-save-email-attachments
Find Gmail messages with attachments and save them to a Google Drive folder.
recipe-draft-email-from-doc
Read content from a Google Doc and use it as the body of a Gmail message.