shortlist-review
Skill12 starsChanged 4 days agoMITView on GitHub
Refine and prioritize vacancies already in the conversation or a saved Scout search using the user's latest preferences. Use for shortlist follow-ups and comparisons; reuse existing results before starting a new search.
In this repository5 other files it gives its agents.
- company-briefSkill
- fit-reviewSkill
- job-researchSkill
- job-search-briefSkill
- linkedin-search-planSkill
The file
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 7
Keep the conversation and its saved search identity.
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Shortlist Review
Keep the conversation and its saved search identity. Incorporate the user's latest refinement without discarding earlier constraints unless they change them. Distinguish supported filtering of known fields from judgments needing more evidence.
If Scout is connected, inspect the discovered refine_shortlist schema. It takes search_id, filters, optional candidate, and optional limit. filters supports include_terms, exclude_terms, companies, exclude_companies, locations, workplace and easy_apply_only. In newer contracts company filters match the employer, and explicit description evidence can supply workplace facts. Candidate retains skills, target titles, locations, workplace and requires_sponsorship; newer servers additionally accept explicit experience_years, target_seniority, remote_from_country and domains. Scout v0.3 supports the optional student facts is_student, degree_completed, graduation_year, fields_of_study, placement_year_available, completed_study_years and grade_classification. Supply only supported fields and evidenced facts, never CV text or contacts. Refinement remains local without a paid provider run.
Company size, salary, sponsorship policy, funding and agency status are not structured filter fields. For those requests, read available vacancy/company evidence and explain the rule used. Keep unknown cases separate unless the user requests their strict exclusion; say how many were excluded for missing evidence. Do not guess headcount or classify an agency solely from an ambiguous name.
An empty workplace-filter result is not proof that no remote roles exist: a source can leave workplace unknown while the description states a remote policy. Inspect those descriptions and present evidence-based findings separately from the structured filter, keeping country restrictions and unconfirmed cases explicit.
If a refinement requires vacancies outside the saved set, explain that and propose a new bounded search in chat. Do not call prepare_search or search_jobs merely to change a local view. start_uncertain on an earlier search means it may already have incurred a provider charge; do not silently create a new plan to bypass it. An error or incomplete result set is not evidence that no suitable jobs exist.
Prioritize with evidence from requirements, CV statements and explicit preferences. For student searches, use program relevance and eligibility checks separately: an appropriate placement can outrank a superficially similar experienced role without being declared eligible. Keep unknown grades, completed study years, placement-year permission and sponsorship unresolved. A start year is not a graduation cohort. Preserve employer/publisher/recruiter attribution and unknown employer identity. Do not re-label a retained provider publication claim or previous source check as newly verified freshness. Scout's deterministic scores are heuristic aids, not a complete assessment or interview probability. A supplied skill absent from a posting is not a missing candidate qualification. Keep observation time and verification limits visible; saved jobs may have changed or closed.
Show the revised shortlist directly in chat, with the reason for changes, known exclusions and important unknowns. Preserve original links and avoid presenting synthetic examples as live vacancies. If Scout is absent, apply the same reasoning to the vacancies already in the conversation; no manual report handoff is required. Do not apply, message, monitor or ask for credentials.
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.

