gtm-engineer-starter

In the author's own words, from each skill's SKILL.md.

• calibrated-judge

  Score rows with Jev, TypeSafe's calibrated evaluation model, and get a probability plus one of four actions (accept, reject, review, insufficient) per row instead of a prose opinion. Use this whenever the user wants to score a list for ICP fit or persona fit, decide which rows a human must review, or check a yes/no claim about a company or person from an evidence packet. Reads the packets that sk…

  skills/calibrated-judge/SKILL.md on GitHub (opens in a new tab)

• crawford

  Build evidence packets for a list of companies or people (Crawford "Job 8"). One packet per row, holding named fields with a source, a date and a character cap on every value, and an explicit "absent, and why" wherever a value is missing. Use this whenever the user wants to score, qualify, or check fit for a list of accounts, wants "the evidence behind each row", or is about to send a list to the…

  skills/crawford/SKILL.md on GitHub (opens in a new tab)

• dossier-builder

  Build per-account customer or prospect dossiers from every system a company has — Gong/calls, CRM, billing, product, support, email — joined into a single sorted timeline with provenance and linking_trust. Use this whenever the user mentions building a dossier, rebuilding an account, running a timeline pipeline, "we're on a Zoom with RevOps and need to walk through their data," consolidating cust…

  skills/dossier-builder/SKILL.md on GitHub (opens in a new tab)

• dossier-data-auditor

  Pre-flight partner to the dossier-builder skill — audit customer/prospect data quality before a dossier build (or after, or standalone). Point it at any folder of customer data (CSV/XLSX/SQLite/Parquet/JSONL); it classifies columns, verifies join keys and company domains, reconciles conflicting revenue columns, and writes an overrides file dossier-builder auto-detects. Launch it whenever dossier-…

  skills/dossier-data-auditor/SKILL.md on GitHub (opens in a new tab)

• wiki-lint

  Lint a project's wiki/ directory (Karpathy-style LLM Wiki + Dietl lifecycle states): catch orphan pages, missing cross-references, stale claims, contradictions, and frontmatter conformance issues. Surfaces problems; doesn't auto-fix substantive content.

  skills/wiki-lint/SKILL.md on GitHub (opens in a new tab)

• win-loss-rewind

  Won/Lost/10X — builds a customer-fit discriminator by working backward from outcome (which customers won, which were lost, and which are 10X, meaning lifetime revenue at least 10 times the median customer's). Takes Gong calls + a list of customers labeled won/lost/healthy/churned, distrusts the user's own labels until verified, auto-segments customers without assuming industry, generates "what wo…

  skills/win-loss-rewind/SKILL.md on GitHub (opens in a new tab)

No skill matches that filter.

Similar skills

• reaby morluto Reverse engineer anything with agents, from app behavior down to native binaries. 291k stars · 1 skill
• ECCby affaan-m The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Cod… 282k stars · 30 skills
• andrej-karpathy-skillsby multica-ai A single CLAUDE.md file to improve Claude Code behavior, derived from Andrej Karpathy's observations on LLM coding pitfalls. 220k stars · 1 skill
• ponytailby DietrichGebert Makes your AI agent think like the laziest senior dev in the room. The best code is the code you never wrote. 165k stars · 6 skills
• ui-ux-pro-max-skillby nextlevelbuilder An AI skill that provides design intelligence for building professional UI/UX across multiple platforms. 137k stars · 7 skills
• graphifyby Graphify-Labs Turn any codebase, with its docs, SQL schemas, configs, and PDFs, into a queryable knowledge graph. A /graphify skill for Claude Code, Curs… 127k stars · 1 skill