Just ask in plain English — your client picks the right tool. Examples below.
| Tool | What it does | Example prompt |
|---|
extract_resume_data | Extract 200+ structured fields (skills, experience, education, contact) from resume text. | "Parse this resume and list the candidate's skills and experience: <paste>" |
extract_resume_data_from_url | Same, but fetches the resume from a public URL (PDF, DOCX, RTF…). | "Parse the resume at this link: https://…/cv.pdf" |
extract_job_data | Extract required skills, experience range, salary and qualifications from a JD. | "Parse this job description and list the required skills: <paste>" |
| Tool | What it does | Example prompt |
|---|
lookup_skill | Authoritative detail on a skill: aliases, related skills, ontology, ONet/ESCO. | "What is the skill Kubernetes? Show related skills." |
lookup_job_profile | Detail on a role incl. the skills required for it. | "Give me the skills to be a QA engineer." |
autocomplete_skill | Typeahead skill suggestions for a partial term. | "Suggest skills starting with 'java'." |
autocomplete_job_profile | Typeahead job-title suggestions for a partial term. | "Suggest job titles starting with 'data'." |
| Tool | What it does | Example prompt |
|---|
redact_resume | Mask/remove bias & PII fields (name, photo, age, contact) for blind hiring. | "Redact name, email and phone from this resume: <paste>" |
reformat_resume_with_template | Restyle a resume into a branded template (PDF/DOCX/…). | "Reformat this resume into template TM003 as a PDF." |
convert_document_format | Convert a document's file format (e.g. DOCX→PDF). | "Convert this resume to PDF." |
tag_entities | Tag job titles, skills, cities, degrees, organizations in free text. | "Extract entities: Senior Java Developer at Infosys, Bangalore." |
extract_contacts | Pull name, email, phone, address from free text. | "Get the contact details from this text: …" |
geolocate | Resolve latitude/longitude for locations. | "Get coordinates for Bangalore and London." |
classify_job_zone | Determine the O*NET Job Zone level (1–5) for a role. | "What O*NET job zone is this candidate? <paste>" |
| Tool | What it does | Example prompt |
|---|
generate_interview_questions | Analyses the role requirements against the candidate's profile and generates a tailored interview kit: a candidate–job match score plus questions across nine categories (technical, behavioral, situational, problem-solving, culture-fit, salary and more). | "Generate interview questions for this candidate against this job. Resume: … JD: …" |
generate_jd | Drafts a complete, publish-ready job description from a job title — seniority level, employment type, role overview, responsibilities, requirements and benefits. Add responsibilities, qualifications, skills or company info for a richer draft. | "Write a job description for a Senior Python Engineer at a fintech startup." |
skill_mapping | Finds every skill mentioned in a resume and maps it to its canonical RChilli taxonomy label, resolving aliases and informal names — standardised candidate data ready for search, matching or analytics. | "Map the skills in this resume to the taxonomy: <paste>" |
map_job_to_skills | Builds the expected skill profile for any job title: categorised technical, soft and domain-specific skills. Useful for writing JDs, screening criteria or competency frameworks. | "What skills should a Data Scientist have?" |
detect_bias | Audits a resume, a job description, or both for biased, exclusive or discriminatory wording (gendered terms, age signals, ableist phrasing) and suggests neutral alternatives — supports fair-hiring review before publishing. | "Check this job description for biased language: <paste>" |
analyze_skill_gap | Compares a candidate resume against a target job description and returns three views: skills present in both, skills the JD requires that the resume lacks, and an overall match score. | "What skills is this candidate missing for this role? Resume: … JD: …" |
recommend_learning_path | Turns a skill-gap analysis into a development plan: an ordered list of learning recommendations, each with a topic, suggested resource types (course, book, article) and an estimated time to completion. | "Recommend a learning path for this candidate to become a DevOps engineer." |
normalize_candidate_data | Cleans up messy HR text from any source — resume snippets, JD fragments, custom records — into one standard structured format with consistent fields, ready for ATS ingestion or further processing. | "Normalise this raw candidate record for my ATS: <paste>" |
match_end_to_end | Runs the full matching pipeline in a single call — extraction, taxonomy normalisation and scoring — returning a ranked, explainable resume-to-job fit with an audit trail. | "Match this resume against this job end to end. Resume: … JD: …" |
screen_application_batch | Runs a multi-turn pre-screening conversation for a candidate against a job description — send the resume and JD first, then answer the agent’s follow-ups using the same conversation ID. | "Pre-screen this candidate for this role. Resume: … JD: …" |