application-setup
Skill139 starsChanged 4 months agoMITView on GitHub
Set or review durable Job Apply browser, application-flow, task model, and Codex reasoning-effort preferences.
In this repository10 other files it gives its agents.
- AGENTS.mdAGENTS.md
- agent-workflowsSkill
- account-setupSkill
- answer-memorySkill
- job-applySkill
- job-preferencesSkill
- job-searchSkill
- job-title-discoverySkill
- job-workspaceSkill
- saved-jobsSkill
The file
Files
Application Setup
Set up or selectively update the owner's durable application-flow and agent model preferences. Read answer-memory first for plugin resolution, Store initialization, private temporary inputs, and exact-revision handling, then read setup questions.
Inspect the current canonical profile before asking questions with
node "<plugin-root>/apps/companion/command.mjs" store profile-inspect. If the owner asks to
view setup, report the saved choices without starting a questionnaire. If the owner
already specified one or more changes, apply those changes and ask only for missing
choices needed to complete the requested setup. Accept partial setup and never erase
an unrelated saved choice.
Save only the changed keys beneath applicationPreferences or agentModelPreferences
with profile-patch, the inspected revision, and source user. Read
agent model preferences
before changing model IDs or Codex reasoning effort. Confirm the stored choices from a fresh
inspection. These preferences guide future agent behavior; they never grant fill,
sensitive-answer, login, account, remember, or final-submission authority.
preferredAutomationMode chooses which mode the application agent should offer. It
is not a live grant. Guided needs no durable authority; Autofill to Review and
Campaign to Review still require the owner to approve the exact current job scope,
expiration, and sensitive-answer references under
application automation.
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
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.

