Resume tailoring agents on Claude Code based on experience bank; renders LaTeX.
Agent Claude Code
Structuring stage of bank onboarding. Reads the ingested raw sources (resumes, notes) listed in the bank's .ingest/manifest.json and writes the two-layer bank per spec/bankformat.md (experiencebank.md for evidence units, profile.json for the record layer), plus .ingest/gaps.md, the ranked interview agenda. Transcribes…
Agent Claude Code
Parse stage of the tailoring pipeline. Receives the application workspace path, reads the posting at /jd.txt, extracts the tracking fields (company, role, ...), and writes /.run/parsedjd.json when absent (reused applications keep theirs). Returns a status, never a path.
Agent Claude Code
Final close-out stage of the resume-tailoring pipeline. Reads the deterministic gate status from the workspace's .build/, completes the workspace's gaps.md with run-level items, and writes the workspace's .run/run.json (the single-run record that code derives everything cumulative from). A frontmatter Stop hook then…
Agent Claude Code
Selects which true bank material to surface for the JD and writes the tailored resume CONTENT in one pass. Reads the raw JD + evidence map + schema, plans the selection from the map, confirms the chosen material against the bank, then writes /.run/resume.json (the draft, against spec/resumeschema.json) and appends…
Agent Claude Code
Independently scores the tailored resume content ( /.run/resume.json) against the quality rubric (.claude/skills/tailor/rubric.md), the bank, and the JD, as a skeptical senior reviewer judging cold. Truthfulness is a hard gate; every other dimension is scored to a defined bar and tagged MATERIAL or MINOR. Writes…
Agent Claude Code
Applies the evaluator's findings to /.run/resume.json as TARGETED field patches instead of regenerating. Applies every MATERIAL fix (grounding never dismissed, never fabricated), judges each MINOR (apply or drop with a logged reason), writes the patched resume.json + a fix log. Does not re-select, re-draft from…
Agent Claude Code
Audits the fix pass of the tailoring pipeline. Verifies every field the fixer changed still traces to the bank and judges each prior finding resolved or not, writes /.run/reviewreverify.md ending in VERDICT: PASS or VERDICT: REVISE, and returns the verdict. Runs cold; sees the files, not anyone's reasoning.