# For AI agents `hired` ships artifacts for coding agents alongside its code. This page lists them, says where each lives in the repository, and points at the machine-readable copies of this documentation. ## Skills Skills are folders holding a `SKILL.md` (the [Agent Skills](https://agentskills.io) format): a description that tells an agent when to use it and a body with the procedure. Install one into your agent with `gh skill` (any host: `--agent claude-code`, `copilot`, `cursor`, `codex`, `gemini`), or use the copy bundled in the wheel. ### `hired-align` Use to analyze how well a candidate fits a job description and to accumulate reusable knowledge about the candidate. Triggers on "analyze this JD against my profile", "how well do I match this job", "what are my gaps for this role", "help me respond to a recruiter about this job", "align my profile to this job description", or when given one or more JDs to assess for a candidate. Classifies each requirement into strong-match / adjacent-transferable / gap-learnable / gap-hard, asks clarifying questions instead of assuming gaps, and persists everything (facts + Q&A + reports) outside the repo under ~/.local/share/hired/. ```bash gh skill install thorwhalen/hired hired-align --agent claude-code ``` Source: [`skills/hired-align`](https://github.com/thorwhalen/hired/tree/HEAD/skills/hired-align). ### `hired-dev` Use when developing or modifying the `hired` package itself — adding a renderer or job-search source, working on the resume pipeline / AI agents (`tools.py`, `content.py`, `resume_agent.py`), the workflow helpers, or hired's dependencies/CI. Triggers on editing files under the hired repo, "add a job source to hired", "add a renderer", "wire the AI agent", or changing hired's extras. ```bash gh skill install thorwhalen/hired hired-dev --agent claude-code ``` Source: [`skills/hired-dev`](https://github.com/thorwhalen/hired/tree/HEAD/skills/hired-dev). ### `hired-interview-prep` Use to prepare a candidate for a headhunter call or company interview. Triggers on "prep me for the interview", "I have a call with the recruiter/headhunter", "help me get ready for the Socure interview", "what should I study before the interview", "research the company before I talk to them". Refreshes the alignment report from the latest Q&A, researches the company and people, and produces study briefings on the subjects/terminology the candidate is weakest on — each anchored by analogy to the candidate's own experience. ```bash gh skill install thorwhalen/hired hired-interview-prep --agent claude-code ``` Source: [`skills/hired-interview-prep`](https://github.com/thorwhalen/hired/tree/HEAD/skills/hired-interview-prep). ### `hired-usage` Use when helping a job seeker with `hired` — tailor/generate a resume and render it to PDF/HTML, search jobs across boards, score/match a resume to a job, run an ATS-compatibility check, draft a cover letter, or track applications. Triggers on "tailor/build a resume", "render resume to PDF", "search for jobs", "match resume to job", "ATS check", "cover letter", "track applications". Core install `pip install hired`; features need extras (`hired[pdf]`, `hired[search]`, `hired[ai]`, `hired[all]`). ```bash gh skill install thorwhalen/hired hired-usage --agent claude-code ``` Source: [`skills/hired-usage`](https://github.com/thorwhalen/hired/tree/HEAD/skills/hired-usage). ## Subagents Subagents are Markdown files with a frontmatter (`name`, `description`, `tools`) and a system prompt as the body. Copy one into your project's `.claude/agents/` (or your agent host's equivalent) to delegate the task it describes. ### `hired-alignment-report` Generates a complete JD↔candidate alignment report from ALL available evidence (candidate facts, Q&A, company report) — deliberately WITHOUT reading any existing alignment report, so it is an unanchored, reusable generator. Use to create a report for a new JD, or as the "fresh second opinion" in a heavy alignment review. Source: [`.claude/agents/hired-alignment-report.md`](https://github.com/thorwhalen/hired/tree/HEAD/.claude/agents/hired-alignment-report.md); tools: `Read, Bash, Grep`. ### `hired-alignment-review` Reviews and refreshes an existing JD↔candidate alignment report against current evidence, in light (default) or heavy mode. Light mode patches false negatives the newer Q&A now resolves; heavy mode regenerates a fresh report from all evidence and adversarially diffs it against the existing one. Always proposes changes for approval — never applies silently. Source: [`.claude/agents/hired-alignment-review.md`](https://github.com/thorwhalen/hired/tree/HEAD/.claude/agents/hired-alignment-review.md); tools: `Read, Bash, Grep`. ### `hired-company-research` Researches a hiring company (and the relevant people) for interview/headhunter prep — business, products, market position, recent news/funding, the data-science/AI org, competitors, culture, and likely interview focus. Produces a structured company report and persists it to the candidate knowledge base. Source: [`.claude/agents/hired-company-research.md`](https://github.com/thorwhalen/hired/tree/HEAD/.claude/agents/hired-company-research.md); tools: `WebSearch, WebFetch, Read, Bash, Grep`. ### `hired-interview-prep` Produces interview-prep study briefings on the subjects and terminology a candidate should master for a job — focused on what they know LEAST, but every new concept anchored by analogy to the candidate's OWN experience, concepts, and jargon. Uses the alignment report (to target gaps), the candidate knowledge base (to find anchors), and web research. Persists briefings. Source: [`.claude/agents/hired-interview-prep.md`](https://github.com/thorwhalen/hired/tree/HEAD/.claude/agents/hired-interview-prep.md); tools: `WebSearch, WebFetch, Read, Bash, Grep`. ### `hired-profile-ingest` Extracts atomic, provenance-bearing facts about a candidate from raw documents (CVs, bios, resumes, publication lists, LinkedIn exports) for the hired candidate knowledge base. Give it file paths; it returns structured fact records. Read-only on documents; never writes to the repo. Source: [`.claude/agents/hired-profile-ingest.md`](https://github.com/thorwhalen/hired/tree/HEAD/.claude/agents/hired-profile-ingest.md); tools: `Read, Bash, Grep, Glob`. ### `hired-requirement-analyst` Classifies one or more job-description requirements against a candidate's known facts for the hired alignment analyzer. Given requirement text plus relevant candidate facts/synopsis, returns honest per-requirement judgments (match state, match type, level, evidence, closeability, AI-leverage) as structured records. Use to parallelize analysis of a large JD. Source: [`.claude/agents/hired-requirement-analyst.md`](https://github.com/thorwhalen/hired/tree/HEAD/.claude/agents/hired-requirement-analyst.md); tools: `Read`. ## Instruction files Files agents read before working in this repository. - [`.claude/CLAUDE.md`](https://github.com/thorwhalen/hired/tree/HEAD/.claude/CLAUDE.md): read by Claude Code ## Machine-readable documentation This site publishes the same documentation in forms that fit an agent's context window: - [`llms.txt`](https://thorwhalen.github.io/hired/llms.txt): an index of every page with a one-line description ([llms.txt](https://llmstxt.org) format) - [`hired.md`](https://thorwhalen.github.io/hired/hired.md): the whole documentation as one Markdown file - `.html.md`: a rendered Markdown twin of every page, advertised from each page's `` with `` - [`objects.inv`](https://thorwhalen.github.io/hired/objects.inv): the Sphinx inventory: a symbol-to-URL index (`sphobjinv convert plain objects.inv -`)