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 skills add krzysztofsurdy/code-virtuoso --skill retrogit 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/skills/krzysztofsurdy/code-virtuoso/retro)<a href="https://agentmods.dev/skills/krzysztofsurdy/code-virtuoso/retro"><img src="https://agentmods.dev/badge/skills/krzysztofsurdy/code-virtuoso/retro/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/skills/krzysztofsurdy/code-virtuoso/retro"><img src="https://agentmods.dev/badge/skills/krzysztofsurdy/code-virtuoso/retro.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 | $0.00069 | $0.01339 |
| Opus 5.5 | $0.00028 | $0.00536 |
| Sonnet 5.5 | $0.00014 | $0.00268 |
| Haiku 4.5 | $0.00007 | $0.00134 |
Grade A, and why
retro 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 today.
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 — 65 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Retro
An agent session ends and the lessons go with it. The same correction gets typed again next week, and the usual fix, one more line in a rules file, is the weakest one: the file grows, the agent skips lines, and the mistake returns. This playbook looks back at a session, finds where it lost time or quality, and picks for each finding the cheapest fix that lasts. Default to a deterministic check. Write a prose rule only when no check can catch the problem.
This is for the agent's working environment, not a team ceremony. For a Scrum sprint retrospective use the scrum skill.
A repeated mistake with no guardrail is itself a finding. Fixing the code once does not count as a fix.
Procedure
-
Gather the friction. Work from evidence, not a feeling that it went badly:
- The current conversation: corrections the user made, errors, retries, dead ends.
git logandgit difffor the session: reverted edits, fix-up commits, files touched twice.- Command output: failing checks, slow steps, commands that were run and changed nothing.
- Progress or handoff notes if they exist.
If the session is not in front of you, ask the user for the three or four moments that hurt most. Do not invent findings. Seven real ones beat twenty guesses. Scan the categories in categories so nothing obvious is missed.
-
Write each finding as one fact. What happened, how many times, what it cost (turns, retries, a wrong commit). "Agent imported from the old module path in 3 files, user corrected it twice" is a finding. "Agent was sloppy" is not.
-
Classify the fix. Take the first row that fits.
The problem is Fix Lands in A fixed pattern a program can recognise: banned API, import shape, file location, naming, formatting, a missing step before commit Deterministic check Linter or static-analysis rule, formatter config, pre-commit hook, agent hook, CI job, or a test A call that needs taste or context a program cannot judge One short rule The project's rules file, or a path-scoped rule if it applies to one area The agent could not find something, or loaded too much Navigation or context fix A pointer in the rules file, a doc moved or renamed, an instruction file trimmed or split A step that changed nothing, or a tool call repeated Remove or cache it Drop the step from the workflow, script, or skill, or record the answer where the agent finds it
What ships with it
2 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.
- today First seen · 65 lines · 69 tokens per session scan A dc86754df4a4
retro is a skill published in the GitHub repository krzysztofsurdy/code-virtuoso (36 stars, last pushed yesterday), licensed MIT. It adds 69 tokens to every session and 1,339 once invoked, about $0.0003 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 2026-10-24.
Other skills, from other repositories
diagnosing-bugs
A structured process for diagnosing difficult software bugs and performance slowdowns. It starts by creating a repeatable pass-or-fail check, such as a failing test, HTTP request, command, or browser script.
incident-postmortem
Escribe un postmortem de incidente estructurado o revisión post-incidente. Úsalo cuando se te pida escribir un postmortem, reporte de incidente, revisión P1/P2, reporte de apagón o RCA (análisis de causa raíz). Produce un postmortem sin culpables con cronología, causa raíz, factores contribuyentes, resumen de impacto…
agent-incident-postmortem
Run a blameless postmortem for an incident caused by an AI agent or LLM feature — hallucinated facts shipped to users, runaway tool use, prompt injection, cost blowouts, or wrong actions taken autonomously. Use when asked to write up an AI incident, analyse why an agent did something wrong, or produce corrective…
agent-design-review
Review an LLM agent design and find where it will be unreliable, expensive, or unsafe. Use when asked to review an agent architecture, critique a multi-step/tool-using agent, debug an agent that loops or goes off-task, or harden an agent before launch. Produces a structured review — task fit, control flow, tools…
prompt-debugging
Figure out why a prompt isn't working and fix it — diagnose the actual failure (ambiguity, missing context, wrong format, conflicting instructions) instead of randomly rewording. Use when asked why isn't my prompt working, the AI keeps ignoring my instructions, my prompt gives inconsistent results, or how do I fix…
prompt-optimizer
Diagnose and rewrite an underperforming LLM prompt so it produces reliable, well-structured output. Use when asked to improve a prompt, fix a prompt that gives inconsistent or wrong results, reduce hallucination/refusals, or make output follow a format. Produces a rewritten prompt with a diagnosis of what was failing…