Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/anton-abyzov/vskill/skill-buildernpx skills add anton-abyzov/vskill --skill skill-buildergit clone --depth 1 https://github.com/anton-abyzov/vskillWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/anton-abyzov/vskill/skill-builder)<a href="https://agentmods.dev/skills/anton-abyzov/vskill/skill-builder"><img src="https://agentmods.dev/badge/skills/anton-abyzov/vskill/skill-builder.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00050 | $0.02200 |
| Opus 5 | $0.00025 | $0.01100 |
| Sonnet 5 | $0.00010 | $0.00440 |
| Haiku 4.5 | $0.00005 | $0.00220 |
Grade A, and why
skill-builder scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 223 lines — stays where its author put it; the contents beside it link to each section on GitHub.
skill-builder
Author new universal skills from natural-language prompts across any agent host (Claude Code, Cursor, Codex, Gemini CLI, Windsurf, OpenCode, and the rest of the 50+ agents in the vskill registry).
What this skill does
skill-builder is a cross-tool authoring meta-skill. When the host agent
sees a trigger phrase (see frontmatter description), this skill takes
over and produces a new SKILL.md — emitted for every target agent, not
just the one currently invoking it. The emitted skills carry an
x-sw-schema-version: 1 marker so future tooling can evolve the format
without breaking older installs.
It does this by delegating to the real generator via a three-path fallback chain:
- Path A —
vskill skillCLI (preferred). The CLI owns the full authoring pipeline: prompt parsing, target-agent lowering, divergence report, security scan, and disk write. - Path B — vskill Studio (browser). When the CLI is not on PATH but the npm package is present (e.g. a project-local install), start the eval server and walk the user through the Studio UI.
- Path C — Anthropic
skill-creator(Claude Code only fallback). Delegate to the built-in Anthropic authoring skill. This produces a single Claude-flavored SKILL.md with no universal targets, so we log the degradation so users know what they're getting.
Each path is tried in order; the first one whose detection check succeeds wins. If none match, we emit a remediation message and stop.
Detection script
The host agent runs this snippet (or the equivalent) before authoring anything. Return on the first path whose probe exits 0.
# Path A: vskill CLI on PATH
if which vskill >/dev/null 2>&1; then
# Invoke the CLI. --targets is optional; default is the full
# universal set. --prompt is the user's natural-language request.
vskill skill new --prompt "<USER_PROMPT>" [--targets=<csv>]
exit $?
fi
# Path B: vskill package resolvable but no CLI on PATH
if node -e "require.resolve('vskill')" >/dev/null 2>&1; then
vskill eval serve
echo "[skill-builder] open the URL printed above and use the Studio 'New skill' flow"
exit 0
fi
# Path C: Claude Code host with Anthropic skill-creator installed
if [ "${SW_HOST_AGENT:-claude-code}" = "claude-code" ] \
&& test -d "$HOME/.claude/skills/skill-creator"; then
echo "[skill-builder] fallback mode — universal targets not emitted; install vskill for universal support"
# Delegate to the Anthropic built-in. The host agent will route
# the original user prompt into skill-creator.
exit 0
fi
# Nothing matched — print remediation and fail loudly.
cat <<'EOF'
[skill-builder] no authoring backend available. Install one of:
npm install -g vskill # preferred — universal targets
claude plugin install skill-creator # Claude Code only, no universal emit
EOF
exit 1
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 6d ago First seen · 223 lines · 50 tokens per session scan A 4f4c4e6f94b4
skill-builder is a skill published in the GitHub repository anton-abyzov/vskill (45 stars, last pushed 3d ago), licensed MIT. It adds 50 tokens to every session and 2,200 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
optimize-skill-loop
Deterministic outer-loop optimizer for a skill that already carries a frozen eval (produced upstream by an eval source such as create-skill-eval). Runs the frozen eval across user-chosen agents on fresh headless sessions via a pluggable adapter, grades quality/cost/perf, keeps the pareto-best variant, and versions…
create-skill-eval
Author a frozen, graded eval harness for ONE inner skill and freeze it inside that skill as a conformant eval/ directory (scenarios + deterministic evaluator + rubric + fixtures + HARNESS.md manifest). The eval content is skill-specific; this skill is the reusable methodology that produces it. Its output is consumed…
prereview
Review the current diff against this team's conventions and invariants before it ships. Invoke with /prereview before opening a PR, or run as a pre-push gate.
fork-and-publish-skill
Customizes an installed agent skill and ships it back upstream or as a versioned fork using qvr's edit/publish authoring loop. Use when a user wants to modify, customize, fork, release, or publish a qvr skill — e.g. "edit this skill", "publish my changes", "fork a skill to my own repo", "cut a v1.0.0 release of a…
onboard-skills
Discovers and installs agent skills into a project (or the user-global lane) with the qvr CLI, where qvr.toml declares intent and qvr.lock is the resolved source of truth for what loads. Use when a user wants to find, add, register, or install skills from a skills registry or GitHub repo with qvr — e.g. "register a…
reproduce-skill-env
Reproduces an exact agent-skill set across machines, teammates, and CI using qvr's portable manifest and lockfile guarantees. Use when a user wants to share, pin, replicate, or CI-gate their qvr skills — e.g. "export my skills", "import this skill manifest", "pin everything to exact commits", "make skills…