saved-jobs
Skill139 starsChanged 4 months agoMITView on GitHub
Retrieve and present the user's canonical saved jobs from the Job Apply Store without changing them.
In this repository10 other files it gives its agents.
- AGENTS.mdAGENTS.md
- agent-workflowsSkill
- account-setupSkill
- answer-memorySkill
- application-setupSkill
- job-applySkill
- job-preferencesSkill
- job-searchSkill
- job-title-discoverySkill
- job-workspaceSkill
The file
Files
Saved Jobs
Read answer-memory for installed root resolution and the QA-route rule. Before listing, run store paths through the same command router and check whether its returned Store root directory exists. If it does not, report that there are no canonical records to retrieve; do not initialize the Store for a read request. When it exists, use the canonical TypeScript command router:
node "<plugin-root>/apps/companion/command.mjs" store job-list --status savedThis shows active canonical records whose status is saved; it excludes trashed records. Do not read Store files directly or substitute LinkedIn's Saved Jobs page, search reports, or the legacy Markdown queue. Treat corrupt or unavailable Store data as an error, never as an empty list. The command router may perform its normal activation for an existing Store; this skill makes no job mutation.
For company, role, location, source, or text filters requested by the owner, filter the returned records in memory; job-list does not provide those selectors. State the filters and matching count. Present each match's role, company, location or work arrangement, source, URL, and last update when present. Mark absent fields as unknown, preserve the exact canonical job ID for follow-up, and state the number returned. Avoid printing descriptions, notes, provenance, resume references, or applicant data unless the owner specifically asks for a relevant field. If the user asks for another active status, use job-list --status <status> and label it accurately. Trash is not a status: only when explicitly requested, use job-list --trashed-only for trashed records or job-list --include-trashed for active and trashed records; add --status saved if the owner wants that status within the requested trash scope.
This skill only retrieves and presents records. It never searches sites, changes status, edits jobs, previews or commits queue imports, opens application forms, or launches a worker. For a selected application, hand the exact canonical ID to Job Apply; for a new opportunity search, use Job Search.
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.

