Build or update the JobClaw master profile — the single file every other job-hunt skill reads. Ingests a resume (extracting every embedded URL), LinkedIn PDF, exported AI-memory, and links, then asks only for what's missing in one batched, optional gap prompt. Captures career-goal context for downstream skills.
Orchestrate the full apply chain for one job — capture the job once, score fit, tailor the resume (PDF), write the cover letter, and answer the application questions — in auto-pilot (run end-to-end, pause only at decision/send gates) or review (confirm each step) mode. Stops before anything that sends or submits. The single entry point for "apply to this job".
Render a resume to a polished PDF via rendercv (Typst engine). Exposes all 9 built-in themes (engineeringresumes is the ATS-safe default) and embeds every link as a real, clickable PDF annotation. Reads the tailor-resume sidecar so the tailored content and links survive intact.
Produce a job-specific, ATS-optimized resume from the user's master profile and a target job — reordering, reframing, and surfacing the most relevant real experience and keywords. Enforces a content rubric (bullet shape, length, parallel grammar, metric coverage), then renders a polished PDF by default (via render-resume) and visually QAs it. Never fabricates.
Answer one or many application-form questions from the master profile, obeying per-type format rules — numeric → number only, yes/no → Yes|No, short → ≤ 2 sentences, paragraph → ≤ max_chars, enum → one of options. Batch JSON in/out (for JobClaw form-fill); single-question paste mode for interactive use. Flags any question without profile evidence rather than fabricating an answer.
Coach-style multi-turn conversation grounded in the master profile. The ONE skill allowed to read AND paraphrase (never quote) the profile's `context:` block — career_goal and additional_info — to direct advice. Last 8 messages of history retained.
Coach a counter-offer — leverage analysis, floor / target / walk-away ranges per comp component, and a draft counter email the human sends (never auto-send). Reads companies/<slug>.json for market comp when present; otherwise asks for an inline range or runs a thin WebSearch. Region pack drives currency / structure (USD base+equity vs INR LPA CTC fixed+variable).
Draft a reply email matched to one of the taxonomy's reply intents — reply-schedule, reply-accept (with sub-templates for scheduling / assessment / offer), reply-decline, reply-ask, reply-info, follow-up. Mirrors inbound tone; interpolates dates / links; uses the profile's signature. Never auto-sends.