👥 HR & Recruiting Prompt Generator Free · No Signup

Boolean Sourcing String Generator — Validated Strings for 4 Platforms

⚡ Quick Answer

A Boolean sourcing string combines job titles, skills, and exclusions with AND, OR, and NOT operators to find candidates who match a brief. This generator builds strings for LinkedIn Recruiter, LinkedIn search, Google X-ray, and GitHub — then validates each one for balanced parentheses, closed quotes, and dangling operators before you run it.

⚡ Boolean Sourcing String Generator — Validated Strings for 4 Platforms

Configure your facts below to generate a production-ready, constraint-locked prompt.

100% Client-Side

Why this page has to beat a tool — and how it does

Boolean is the one task in HR where a free working generator already ranks: paste a role brief in, get strings out. So this page matches that bar and then goes past it in three ways a string-only tool doesn't:

  1. You get the prompt, not just the string. Every output includes the reusable AI prompt that generated the strings, tuned with your titles, skills, and exclusions — so next week, for a different role, you can regenerate and adapt inside ChatGPT, Claude, or Gemini without coming back to any tool.
  2. Three widths, not one string. Broad (maximum pool), semi-detailed (the working string most recruiters should start from), and detailed (tight, senior, exclusion-heavy) — because the right way to source is to start wide, read results, and narrow deliberately.
  3. Validation is explained, not just performed. Each string is checked client-side for the errors that silently wreck searches — and the page shows you the rules, so you can hand-fix a string in seconds when you're editing inside LinkedIn at midnight.

How Boolean sourcing works (the 60-second version)

A Boolean string has three building blocks:

Quotes force exact phrases ("product manager", not product AND manager anywhere on the page). Parentheses control the order of operations, exactly like arithmetic. Get either wrong and the string still "works" — it just quietly returns the wrong people. That is why validation matters more than generation.

The validation rules (checked on every string this page generates)

  1. Balanced parentheses. Every ( has its ). An unclosed group makes LinkedIn Recruiter reject the string — or worse, regroup your logic.
  2. Closed quotes. Every opened " is closed. One missing quote turns the rest of the string into a single phrase search and returns near-zero results.
  3. No dangling operators. No string may begin or end with AND, OR, or NOT, and no two operators may sit side by side (AND OR).
  4. Uppercase operators. AND / OR / NOT must be uppercase to be read as operators on LinkedIn; lowercase "and" is searched as a word.
  5. One concept per group. Title synonyms in one group, skills in the next, exclusions in NOT groups — mixed groups are where logic errors hide.
  6. Platform limits respected. The LinkedIn search bar accepts shorter, simpler strings than Recruiter; Google X-ray strings must carry a site: operator and use - for exclusions; GitHub search uses qualifiers (language:, location:, followers:) alongside keywords, not classic Boolean alone. The generator reformats per platform instead of pasting one string everywhere.

The prompt template

WORKED EXAMPLE / CODE 
You are an expert technical sourcer who writes precise Boolean search strings.

BRIEF:
- Target role: [job title]
- Title synonyms / variants (include seniority forms): [list]
- Must-have skills — a candidate lacking any group is out: [group skills, e.g. Group A: Python; Group B: Django OR Flask]
- Nice-to-have skills: [list]
- Exclusions — titles, employers, seniority levels, or fields to remove: [list]
- Locations to include: [list] / Locations to exclude: [list]

BUILD THREE VERSIONS — Broad, Semi-detailed, Detailed — for EACH platform below:

1. LINKEDIN RECRUITER: full Boolean. Structure: (title OR group) AND (must-have group) AND (must-have group) NOT (exclusion group). Exact phrases in quotes; operators uppercase.
2. LINKEDIN SEARCH BAR: a shortened version — title group + top two must-haves only, no nested groups deeper than one level.
3. GOOGLE X-RAY: begin with site:linkedin.com/in (or the regional domain I specify), then keywords; exclusions with a minus sign (-term); no NOT operator; keep under 32 words.
4. GITHUB (for engineering roles): combine keyword terms with qualifiers — language:[primary language] location:[location] followers:>= [threshold if I gave one]; if the role is not an engineering role, write "Not applicable" for this platform and say why.

RULES FOR EVERY STRING:
- Synonyms belong in OR groups; never put two different concepts in one group.
- Quote every multi-word title and skill.
- Every parenthesis balanced, every quote closed, no operator at the start or end of a string, no doubled operators.
- After each string, add one line: "To widen: …" naming the single group to loosen, and "To narrow: …" naming the single group to tighten. Change one block at a time — that is how sourcing iteration works.
- If my brief is missing something that changes the string materially (e.g. no location), ask me in one line at the top instead of assuming.

NEGATIVE CONSTRAINTS:
- Do NOT include any term that targets or proxies a protected characteristic — no gendered terms, no age signals ("recent graduate" as a youth proxy, "digital native"), no nationality or language-of-origin terms unless the role itself requires the language for the work, in which case use the language name only.
- Do NOT invent skills the role "probably" needs. Use only my brief.
- Do NOT add employer names to the positive groups unless I listed them as targets.
Best model for this: GPT-6.1 (strict operator logic and format discipline). Claude Sonnet 5.5 is the better choice when the brief is messy prose that needs interpreting first.

Worked example

Filled inputs: Target role: Senior Backend Engineer. Synonyms: "Back-end Developer", "Server-side Engineer", "Software Engineer (Backend)". Must-haves: Group A: Python; Group B: Django OR Flask; Group C: PostgreSQL. Nice-to-haves: AWS, Kubernetes. Exclusions: intern, student, front-end, data engineer. Location: Bengaluru or remote-India.

Generated strings (semi-detailed width):

LinkedIn Recruiter:

WORKED EXAMPLE / CODE 
("senior backend engineer" OR "senior back-end developer" OR "senior server-side engineer" OR "backend engineer") AND (python) AND (django OR flask) AND (postgresql) NOT (intern OR student OR "front-end" OR frontend OR "data engineer")
Google X-ray:
WORKED EXAMPLE / CODE 
site:linkedin.com/in "backend engineer" python django OR flask postgresql (bengaluru OR bangalore OR india) -intern -student -"front end"
GitHub:
WORKED EXAMPLE / CODE 
language:Python location:Bangalore followers:>=10
Validator output: Parentheses balanced ✓ · Quotes closed ✓ · No dangling operators ✓ · Operators uppercase ✓. To widen the Recruiter string: drop the PostgreSQL group first — it is the rarest term on profiles. To narrow: add "system design" to a new AND group.

Guardrails

Frequently Asked Questions

Practical answers covering model choice, workflow integration, and compliance rules.

It converts a vague brief ("find me a backend engineer") into explicit logic: these titles, these skills, not these. That precision cuts the two wastes of keyword searching — irrelevant profiles that mention the right words in the wrong context, and relevant profiles missed because they use a synonym you didn't type. It also makes a search repeatable and shareable: a colleague can run your exact string and get your exact pool.

Boolean search is deterministic: you define the logic, the database executes it literally, and the same string always returns the same pool. AI sourcing tools infer similarity — they expand your brief with terms you didn't choose and rank profiles by predicted fit. The trade-off is control versus discovery. Boolean is auditable (you can defend exactly why a profile appeared); AI ranking is a black box, which matters when a hiring decision is challenged. Many recruiters use AI sourcing to explore, then Boolean to build the defensible shortlist pool.

Usually one of four causes: a missing quote, so "product manager" matches any page containing both words separately; a concept error, where an OR group mixes titles with skills and the AND logic collapses; skills that appear in "interests" or old jobs rather than the current role; or an exclusion list that was never built, so interns, students, and adjacent roles flood the results. The fix is iterative: read the first screen of results, identify the single pattern causing the noise, and change one block.

Work in this order: (1) list every title variant, including seniority forms, in one quoted OR group; (2) group must-have skills — one AND per genuinely separate requirement; (3) build the NOT group (wrong seniority, adjacent roles, employers you can't approach); (4) add location; (5) validate — balanced parentheses, closed quotes, uppercase operators, no dangling AND/OR/NOT. Then run it, read the first 20 results, and adjust one block at a time. The template on this page encodes exactly this workflow.

Run the five checks in the validation section above: count parentheses in pairs, count quotes in pairs, read the first and last tokens (neither may be an operator), confirm operators are uppercase, and confirm each group holds one concept. The validator on this page performs all five in your browser the moment a string is generated, so a string that reaches LinkedIn has already passed.

Internal links

---