llm-job-pipeline: Command for Claude Code

.claude/commands/jobs-new.md

jobs-new is a command for Claude Code from ncalavera/llm-job-pipeline. It costs 0 tokens per session (2,697 once invoked), scanned A, original, MIT.

A daily command that runs a job-search pipeline in a fixed order: checking data, fetching and filtering vacancies, scoring them, deciding outcomes, and publishing results.

In plain words
What is it for?
Running the daily vacancy workflow while supplying human judgment where the pipeline pauses.
Why use it?
It removes the need to manually coordinate each stage and records progress through checkpoints and approval gates.

Command for Claude Code

Written for Claude Code: installed under .claude/. Also seen: mentions subagents.

This is ncalavera/llm-job-pipeline's own configuration. It tells Claude Code how to work on llm-job-pipeline itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything llm-job-pipeline configures →

Needs its repository: it runs a file that does not travel with it, so clone the repository first. The line is python3 scripts/run_daily.py.

Reuse

Borrowing it

Nothing to install: this file belongs to ncalavera/llm-job-pipeline. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/ncalavera/llm-job-pipeline/main/.claude/commands/jobs-new.md
Clone the repo
git clone --depth 1 https://github.com/ncalavera/llm-job-pipeline

Made for: Claude Code.

Wrote 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.

agentmods badge for jobs-new

README.md
[![agentmods](https://agentmods.dev/badge/commands/ncalavera/llm-job-pipeline/jobs-new/github.svg)](https://agentmods.dev/commands/ncalavera/llm-job-pipeline/jobs-new)
Your own site
<a href="https://agentmods.dev/commands/ncalavera/llm-job-pipeline/jobs-new"><img src="https://agentmods.dev/badge/commands/ncalavera/llm-job-pipeline/jobs-new/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.

agentmods 80×15 button for jobs-new

Your own site · 80×15
<a href="https://agentmods.dev/commands/ncalavera/llm-job-pipeline/jobs-new"><img src="https://agentmods.dev/badge/commands/ncalavera/llm-job-pipeline/jobs-new.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,697 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00000 $0.02697
Opus 5 $0.00000 $0.01349
Sonnet 5 $0.00000 $0.00539
Haiku 4.5 $0.00000 $0.00270

Measured 2d ago against content hash ba42816e78f3, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

jobs-new scanned grade A with 1 finding 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 2d 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.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

probing the public ATS APIs with `curl`
.claude/commands/jobs-new.md · 216 lines

How it starts

The opening of the file, as written. The whole thing — 216 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/jobs-new

One command a day. The deterministic orchestration — stage ORDER, batching, checkpoints, the heartbeat, and the publish gate — lives in scripts/run_daily.py, NOT in this file. You cannot run the stages out of order, because you do not drive them; the driver does. Your job is only the JUDGMENT it pauses for. Works with any coding agent that runs shell and follows this file: the driver does the Python; you prepare evidence at its gates. Human keep/put-aside decisions happen in the dashboard.

Reply in the user's product language. Before you say anything to the user, read the ## OUTPUT_LANGUAGE section of config/user_profile.md (resolve it with python3 -c "import sys;sys.path.insert(0,'scripts');import product_language as p;print(p.resolve())"en/ru). Write ALL your chat, gate summaries and progress notes in that language. The driver already prints its banner/summary in it; match it. (Scoring/verdict data stays as the pipeline emits it — you are translating your own words, not the DB.)


The loop

  1. Launch the driver in the background and show the live card. A foreground command's stdout is invisible until it exits, so a long fetch looks frozen. Launch with run_in_background: true, then every ~20–30s run python3 scripts/run_card.py and post its one-line output to chat (fetch ▕███░░░▏ 18/40 · LinkedIn · +12 · 6m02s). Pace polls with the wait / Monitor primitive — never a foreground sleep (blocked).

    python3 scripts/run_daily.py
    

    Boards you have enabled (via /jobs-add board or python3 scripts/sources.py enable-board <id>) fetch automatically every run — nothing to pass. --boards "a,b,c" adds more boards for THIS run only, unioned ON TOP of the persisted set; it never has to be repeated to keep an enabled board on. See what's enabled with python3 scripts/sources.py (catalogue in docs/job-boards-catalogue.md).

    Not sure which boards fit? python3 scripts/sources.py recommend proposes the ones that match your profile (target field, roles, geography) — an engineer is proposed engineering boards, not someone else's fixed set. It only suggests; enable what you want with enable-board. Boards never auto-enable.

Read the full file on GitHub · 216 lines

Changes

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.

  1. 2d ago Changed · -54 lines ba42816e78f3
  2. 7d ago Changed · +2 lines 3b2041cf51b4
  3. 12d ago First seen · 268 lines · 0 tokens per session scan A 7e21683cfaeb

Subscribe to this mod's changes

jobs-new is a command published in the GitHub repository ncalavera/llm-job-pipeline (2 stars, last pushed 3d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 2,697 tokens. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.