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.
git clone --depth 1 https://github.com/krzysztofsurdy/code-virtuosoWrote 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/commands/krzysztofsurdy/code-virtuoso/help)<a href="https://agentmods.dev/commands/krzysztofsurdy/code-virtuoso/help"><img src="https://agentmods.dev/badge/commands/krzysztofsurdy/code-virtuoso/help/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/krzysztofsurdy/code-virtuoso/help"><img src="https://agentmods.dev/badge/commands/krzysztofsurdy/code-virtuoso/help.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00016 | $0.00479 |
| Opus 5.5 | $0.00006 | $0.00192 |
| Sonnet 5.5 | $0.00003 | $0.00096 |
| Haiku 4.5 | $0.00002 | $0.00048 |
Grade A, and why
help 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 2d 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.
What it actually says
loop-virtuoso
loop-virtuoso runs a deterministic, backlog-driven autonomous execution loop. It feeds an agent one small backlog item at a time. An item flips to verified only when a script runs its literal verifyCommand, that command exits 0, and a write-boundary guard confirms no protected file was touched - never on the agent's say-so.
Two modes
- Mode A - session-supervised (
/loop-virtuoso:start): runs inside your current Claude Code session via aStophook. Best for watching it work, stepping away briefly, and backlogs of roughly 5-15 items. - Mode B - unattended batch (
scripts/run-batch.sh path/to/backlog.json): a freshclaude -pprocess per iteration, with a cumulative cost cap. Best for large backlogs and overnight/CI runs. It is intentionally not a slash command because it spawns long-running detached processes.
Commands
| Command | What it does |
|---|---|
/loop-virtuoso:start [backlog] |
Start Mode A over the backlog (default .code-virtuoso/loop/backlog.json at the git root). Needs a clean working tree and refuses to run on the default branch. |
/loop-virtuoso:status [backlog] |
Report counts, iteration, stall/violation counts, and the progress-log and events tails. |
/loop-virtuoso:cancel |
Stop the loop and record a cancelled event; backlog state is preserved. |
/loop-virtuoso:help |
Show this overview. |
Getting a backlog
You need a .code-virtuoso/loop/backlog.json before starting. Use the backlog-compiler skill to convert a PRD, ticket, or freeform feature description into a loop-ready backlog - it sizes each item to a single context window, orders items by dependency, and writes a literal verifyCommand for each one.
For the full behavior - stop reasons, the write-boundary guard, and configuration - see the iteration-loop skill.
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.
- 2d ago First seen · 28 lines · 16 tokens per session scan A d39a8249dc07
help is a command published in the GitHub repository krzysztofsurdy/code-virtuoso (36 stars, last pushed yesterday), licensed MIT. It adds 16 tokens to every session and 479 once invoked, about $0.0001 per session on Opus 5.5. 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 2026-10-22.
Other commands, from other repositories
a11y-scan
Scan web application components for accessibility violations against WCAG guidelines.
test-prompt
Test an AI prompt against multiple scenarios to verify consistent, quality output.
load-test
Run a load test against an API endpoint to measure throughput and identify breaking points.
test-endpoint
Test an API endpoint with various request scenarios and validate responses.
update-deps
Safely update project dependencies with compatibility verification.
flow-start
Start a new feature, bugfix, or hotfix branch following git-flow conventions.