Agent skill

Aic Version Slots

by ai-dynamo in ai-dynamo/aiconfigurator

A skill your agent uses when adding perf data for a new framework version, deciding which backendversion to query or pin in tests, bumping the maintained (current/previous) versions, retiring old…

Apache-2.0Auto-check passed

Install Aic Version Slots

skills CLI
$ npx skills add ai-dynamo/aiconfigurator --skill aic-version-slots -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install ai-dynamo/aiconfigurator aic-version-slots --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/ai-dynamo/aiconfigurator.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.claude/skills/aic-version-slots .claude/skills/aic-version-slots && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
aic-version-slots
GitHub stars
455
Token cost
~878 tokens
SKILL.md length
425 words
Files
1
Skills in repo
5
Repo updated
First seen
Licence
Apache-2.0

At a glance

A skill your agent uses when adding perf data for a new framework version, deciding which backendversion to query or pin in tests, bumping the maintained (current/previous) versions, retiring old…

  • Works in 3 steps: Collect your op's data under its real… → Do NOT create reuse.yaml markers to make… → Deleting a retired version is a plain…
  • Adding perf data for a new framework version
  • SKILL.md covers The model in one paragraph, As a data contributor…, As a maintainer (systematic… and In tests, plus 1 more section
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Aic Version Slots is an agent skill from ai-dynamo/aiconfigurator. Use when adding perf data for a new framework version, deciding which backendversion to query or pin in tests, bumping the maintained (current/previous) versions, retiring old version directories, or hitting "is not a queryable version" errors from the version-slot gate.

Its SKILL.md is about 880 tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

The repository describes itself as: Offline optimization of your disaggregated Dynamo graph. The licence is Apache-2.0.

When your agent uses it

  • Adding perf data for a new framework version
  • Deciding which backendversion to query
  • Bumping the maintained (current/previous) versions
  • Retiring old version directories

Example prompts

  • “is not a queryable version”
  • “/aic-version-slots”

Workflow steps

3 steps, taken from the first numbered list in SKILL.md.

  1. Collect your op's data under its real version directory
  2. Do NOT create reuse.yaml markers to make old versions queryable, and do
  3. Deleting a retired version is a plain directory deletion, but check the

What it can do on your machine

Read from SKILL.md and the folder at commit f254959. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Aic Version Slots loads about 878 tokens when it runs. Until then it costs about 73 tokens; SKILL.md has 425 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~73
When it runs · the whole SKILL.md, loaded when a task matches
~878

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from ai-dynamo/aiconfigurator at commit f254959, republished under its Apache-2.0 licence (© ai-dynamo). 425 words, ~878 tokens.

Download SKILL.mdSave it as .claude/skills/aic-version-slots/SKILL.md (or your agent's skills folder).
name
aic-version-slots
description
Use when adding perf data for a new framework version, deciding which backend_version to query or pin in tests, bumping the maintained (current/previous) versions, retiring old version directories, or hitting "is not a queryable version" errors from the version-slot gate.

AIC Queryable-Version Slots

The model in one paragraph

Each (system, framework) exposes at most three queryable versions, resolved through aic-core/src/aiconfigurator_core/systems/query_versions.yaml: current (newest maintainer-completed full upgrade; authored), previous (the current before it; authored, may be empty), and next (derived automatically: the highest DATA-BACKED version newer than current anywhere in the fleet — single-op development drops land there). The literal aliases current / previous / next are accepted anywhere a version is requested. Every other version fails loudly. Data directories outside the slots are fill sources, not versions: they keep serving queries through backward fill and the cross-backend whitelist, but have no query entry of their own.

As a data contributor (supporting a new model)

  1. Collect your op's data under its real version directory (data/<system>/<family>/<backend>/<version>/) with full provenance — nothing else needed. If your version is newer than current, it becomes the fleet-wide next automatically; ops your version lacks are served by backward fill from current (never forward).
  2. Do NOT create reuse.yaml markers to make old versions queryable, and do not add per-system version exceptions — frozen baselines (a100, b60) are the only authored overrides.
  3. Deleting a retired version is a plain directory deletion, but check the keep-list first: comm (multi-node curves), wideep, sole-source tables (a table no slot version holds — e.g. kda, raw MLA), dsa small-heads holders, and rust-test fixture coordinates stay.

As a maintainer (systematic upgrade)

Bump current in query_versions.yaml (move the old current to previous) in the same PR that lands the recollected data. The PR's data diff is the review surface: families with new directories were recollected; families without ride backward fill (an accepted approximation). When the previous generation's data is retired later, delete the directories — no markers.

Show full SKILL.md (147 more words)Show less

In tests

  • Product-surface tests (CLI, task, webapp, support matrix) pin slot versions or the aliases. If a slot bump breaks such a test, re-anchor or delete the test — never bypass the gate.
  • Loader-level tests that intentionally address a DATA COORDINATE (strict provenance sidecars, sole-source tables like wideep 0.5.6.post2, the power-measured 0.22 tables, deliberate data-gap coordinates) pass allow_unlisted_version=True with a comment saying which coordinate and why. That flag is a declaration, not an escape.
  • AIC_ALLOW_UNLISTED_VERSIONS=1 exists only as a transition escape for legacy suites and is being retired; do not add new uses.

Error you will meet

ValueError: <backend>/<version> is not a queryable version on <system>; available: {...} — you asked for a non-slot version. Fix the request to a slot/alias; only add allow_unlisted_version=True when you are genuinely addressing raw data coordinates.

Policy authority: the version-slot section of docs/perf_database/collector-v3-op-centric-design.md and .claude/rules/ own the rules; this skill is the procedural summary.

© ai-dynamo, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in .claude/skills/aic-version-slots of ai-dynamo/aiconfigurator.

Open the folder on GitHubat commit f254959

Compare with similar skills

Aic Version Slots next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Aic Version Slots compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Aic Version Slots this skillai-dynamo/aiconfigurator455—~878Automated safety check: PassApache-2.0
Perf ComparisonClickHouse/ClickHouse50k—~3.9kAutomated safety check: NotesApache-2.0
Studio Perfremotion-dev/remotion62k—~370Automated safety check: PassCustom licence
Perf ReportClickHouse/ClickHouse50k—~2.6kAutomated safety check: NotesApache-2.0
BigQuery Slot and Cost Optimizergoogle/skills21k—~2.3kAutomated safety check: PassApache-2.0
Perfhashgraph-online/awesome-codex-plugins1.2k—~3.2kAutomated safety check: PassApache-2.0

Similar skills

  • Perf Comparison

    ClickHouse/ClickHouse

    Evaluate ClickHouse performance test results from existing CI/dashboard data or local perf.py runs.

    50k GitHub stars~3.9k tokensUpdated today
    DatabasesAuto-check: notes
  • Studio Perf

    remotion-dev/remotion

    Official

    Optimize Remotion Studio render performance by minimizing frame-reactive component surfaces, preferring imperative timeline state reads, and memoizing stable render boundaries.

    62k GitHub stars~370 tokensUpdated today
    Media & CreativeAuto-check passed
  • Perf Report

    ClickHouse/ClickHouse

    Analyze CI performance comparison reports for a ClickHouse PR.

    50k GitHub stars~2.6k tokensUpdated today
    DatabasesAuto-check: notes
  • Official

    Analyzes BigQuery slot use, query costs and execution bottlenecks from INFORMATION_SCHEMA to diagnose slow queries, slot contention and unpartitioned scans.

    21k GitHub stars~2.3k tokensUpdated today
    DatabasesAuto-check passed
  • Perf

    hashgraph-online/awesome-codex-plugins

    Make a studio's game run faster with a measured loop, and keep only what really helped.

    1.2k GitHub stars~3.2k tokensUpdated today
    Auto-check passed
  • Perf Labs Perf

    perf-labs/perf

    Performance engineering on x86-64 Linux with the perf-labs/perf toolkit (perf benchmark, perf profile, perf analyze, perf view, perf plot, perf compare, perf info) and system perf.

    110 GitHub stars~5.3k tokensUpdated today
    Auto-check: notes

More from ai-dynamo/aiconfigurator

  • Adapt Server Config

    ai-dynamo/aiconfigurator

    Convert InferenceX DB config/benchmark pairs, Dynamo recipe YAML, or confirmed custom serving configs into validated aiconfigurator estimate requests.

    455 GitHub stars~828 tokensUpdated 20 days ago
    Auto-check passed
  • Aic Collector Op Development

    ai-dynamo/aiconfigurator

    Design, add, review, or modify AIC Collector operations and their case population.

    455 GitHub stars~3k tokensUpdated 20 days ago
    Auto-check passed
  • Aic Auto Collect

    ai-dynamo/aiconfigurator

    A skill your agent uses when running long AIC/aiconfigurator GPU perf auto-collection for a specific GPU/framework/frameworkversion, including draft-PR checkpoints, resumable collectxx.py runs…

    455 GitHub stars~6.9k tokensUpdated 20 days ago
    Auto-check passed
  • Aic Silicon Align

    ai-dynamo/aiconfigurator

    A skill your agent uses when validating AIC predictions against real GPU serving measurements and root-causing fidelity gaps (prediction-vs-measured issues, memory/concurrency mismatches, per-op…

    455 GitHub stars~1.2k tokensUpdated 20 days ago
    Auto-check passed

Questions about Aic Version Slots

What does Aic Version Slots do?

A skill your agent uses when adding perf data for a new framework version, deciding which backendversion to query or pin in tests, bumping the maintained (current/previous) versions, retiring old…. Aic Version Slots is an agent skill from ai-dynamo/aiconfigurator. Use when adding perf data for a new framework version, deciding which backendversion to query or pin in tests, bumping the maintained (current/previous) versions, retiring old version directories, or hitting "is not a queryable version" errors from the version-slot gate.

When should I use Aic Version Slots?

Aic Version Slots fits situations like: adding perf data for a new framework version; deciding which backendversion to query; bumping the maintained (current/previous) versions; retiring old version directories.

How do I install Aic Version Slots in Claude Code?

Run `npx skills add ai-dynamo/aiconfigurator --skill aic-version-slots -a claude-code`. Or copy the skill folder (.claude/skills/aic-version-slots in ai-dynamo/aiconfigurator) into .claude/skills/aic-version-slots in your project. Claude Code loads it when a task matches its description.

How do I install Aic Version Slots in Codex?

Run `npx skills add ai-dynamo/aiconfigurator --skill aic-version-slots -a codex`. Or copy the skill folder (.claude/skills/aic-version-slots in ai-dynamo/aiconfigurator) into .agents/skills/aic-version-slots in your project. Codex loads it when a task matches its description.

Can I use Aic Version Slots in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add ai-dynamo/aiconfigurator --skill aic-version-slots -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/aic-version-slots, .gemini/skills/aic-version-slots, .github/skills/aic-version-slots and .opencode/skills/aic-version-slots in your project.

What does Aic Version Slots need to run?

SKILL.md names no scripts, command-line tools or credentials: Aic Version Slots is instructions for the agent only.

Does Aic Version Slots access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Aic Version Slots safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Aic Version Slots use?

Aic Version Slots is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Aic Version Slots use?

About 878 tokens (SKILL.md is roughly 3.5k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Aic Version Slots?

Skills that share tags, products or a category with Aic Version Slots: Perf Comparison (ClickHouse/ClickHouse, 50k stars), Studio Perf (remotion-dev/remotion, 62k stars), Perf Report (ClickHouse/ClickHouse, 50k stars) and BigQuery Slot and Cost Optimizer (google/skills, 21k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Aic Version Slots?

ai-dynamo (a GitHub organization) maintains it in ai-dynamo/aiconfigurator, which has 455 GitHub stars. The repository holds 5 skills in this directory. The repository was last updated on September 18, 2026.

Source: ai-dynamo/aiconfigurator on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.