agentleFS
qbekaqbeka/jev-job-search/.claude/skills/daily/SKILL.md

daily

Skill368 starsChanged 5 days agoMITView on GitHub

The daily run, which sends a few applications a day with nobody watching, inside the user's standing policy. Use when the user says /daily, with or without a subcommand (setup, dry, run, schedule, stop), "run it every day", "automate this", "schedule it", or asks about the daily limits or data/policy.json.

22 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use monthly and weekly.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • gaiaaiagentSkill

    monthly

    Generate a monthly digest distilling weeklies into trend analysis and narrative arcs

    Also in 3 projects that use this

  • gaiaaiagentSkill

    weekly

    Generate a weekly digest rolling up dailies into higher-level pattern analysis

    Also in 9 projects that use this

The file

We split this skill into the rules it gives an agent

5 rules. 0 are common, 0 are distinctive, and 5 are very specific to this project.

  • 0

    Common

    Most of the 17 edited versions we read keep these.

  • 0

    Distinctive

    This author's own choices. Worth a look.

  • 5

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    If npx jev accounts shows Workday is allowed, a daily run signs in at employers.

    line 27
  2. Very specific:

    If data/profile.json is missing, do /setup first.

    line 33
  3. Very specific:

    Write data/policy.json in the shape of data/policy.example.json, show it to them, and save it only after a yes.

    line 49
  4. Very specific:

    Never raise a limit in DAILY to send more: If they ask for more than 25 a day, say why: no rate is known to be safe, and a board that …

    line 58
  5. Very specific:

    If the permission system declines a command, do not try another way.

    line 61

“Common” is measured: kept by at least 6 in 10 of the 17 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

/daily

The user types a short command; you run the tool for them. They should never have to type npx. docs/DAILY.md is the reference. The daily run is off until the user turns it on.

What the user can type

They typeYou do
/dailyRun npx jev schedule status and read today's data/runs/daily-<date>.json if there is one. Say what is set up and how today went. If data/policy.json does not exist, go to setup.
/daily setupThe questions below, then write data/policy.json after a yes.
/daily dry or /daily dry 3npx jev daily --dry --target <n> (3 when no number). Read the summary with them.
/daily runnpx jev daily. They typed it, so that is their go for today's run. Read the summary back when it ends.
/daily schedule or /daily schedule 08:30npx jev schedule install --at <time> (09:00 when no time). Say: the Mac must be on and signed in at that time, the tool's Chrome window will open, /daily pause removes it again, and /daily stop stops a run that is working.
/daily pausenpx jev schedule remove. No further run starts by itself. Say that a run already working carries on.
/daily stopnpx jev daily --stop. The run that is working finishes the application it is on and opens no other. Say that this does not remove the schedule; /daily pause does.

If npx jev accounts shows Workday is allowed, a daily run signs in at employers. Then you do not start dry or run yourself: show the command in a bash block, which has a Run button, and let them click it.

Setup: the standing policy

If data/profile.json is missing, do /setup first. Then ask, one question at a time:

  • How many applications a day? 15 is the default; 25 is the most.
  • Where may the jobs be? (Vancouver, the rest of Canada, remote, the United States, elsewhere.)
  • Internships, new-grad roles, or both?
  • Is there a fit score below which you would rather not apply? Show the spread in their queue (npx jev queue) so the number means something.
  • Any employers to leave out?
  • A resume written for each job, and a cover letter where a form has a box for one? This takes longer per job.
  • How much may JEV cost in a day? 50 cents is the default; a normal day is a few cents. Say that this limits JEV only: Claude is on their subscription, or on their own API key, and is not counted.

Write data/policy.json in the shape of data/policy.example.json, show it to them, and save it only after a yes. Then offer /daily dry.

What you never do

  • Never start a real run or install the schedule unless the user typed that command (/daily run, /daily schedule) in this conversation. A general "set it up" is not that.
  • Never write or change data/policy.json beyond what the user said.
  • Never raise a limit in DAILY to send more. If they ask for more than 25 a day, say why: no rate is known to be safe, and a board that decides one person is a robot stops taking their applications.
  • If the permission system declines a command, do not try another way. Show the command in a bash block for them to run with one click.

Reading a summary

  • "Waiting for you": a human check or a sign-in only they can pass. They finish it in the tool's Chrome window, then /resume.
  • "Clicked and not confirmed": /resume settles these too.
  • A board left alone until tomorrow asked for a human check. That is the tool being careful, not a fault.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.