AI SDK
vercel-labs/ai-facts
Answer questions about the AI SDK and help build AI-powered features.
Upgrade the pinned upstream AI SDK reference, assess the current Go implementation against it, and register parity work for independent implementation afterward.
OfficialApache-2.0Auto-check passed
Install AI SDK Parity Upgrade
$ npx skills add grafana/ai-sdk --skill ai-sdk-parity-upgrade -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install grafana/ai-sdk ai-sdk-parity-upgrade --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/grafana/ai-sdk.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.agents/skills/ai-sdk-parity-upgrade .claude/skills/ai-sdk-parity-upgrade && rm -rf skills-srcUse ~/.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/
Install the "ai-sdk-parity-upgrade" agent skill from https://github.com/grafana/ai-sdk/tree/main/.agents/skills/ai-sdk-parity-upgrade into .claude/skills/ai-sdk-parity-upgrade/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "ai-sdk-parity-upgrade", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/grafana/ai-sdk/tree/main/.agents/skills/ai-sdk-parity-upgradeType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add grafana/ai-sdk --skill ai-sdk-parity-upgrade -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install grafana/ai-sdk ai-sdk-parity-upgrade --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/grafana/ai-sdk.git skills-src && mkdir -p .agents/skills && cp -r skills-src/.agents/skills/ai-sdk-parity-upgrade .agents/skills/ai-sdk-parity-upgrade && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "ai-sdk-parity-upgrade" agent skill from https://github.com/grafana/ai-sdk/tree/main/.agents/skills/ai-sdk-parity-upgrade into .agents/skills/ai-sdk-parity-upgrade/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "ai-sdk-parity-upgrade", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add grafana/ai-sdk --skill ai-sdk-parity-upgrade -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install grafana/ai-sdk ai-sdk-parity-upgrade --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/grafana/ai-sdk.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/.agents/skills/ai-sdk-parity-upgrade .cursor/skills/ai-sdk-parity-upgrade && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "ai-sdk-parity-upgrade" agent skill from https://github.com/grafana/ai-sdk/tree/main/.agents/skills/ai-sdk-parity-upgrade into .cursor/skills/ai-sdk-parity-upgrade/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "ai-sdk-parity-upgrade", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/grafana/ai-sdk.git --path .agents/skills/ai-sdk-parity-upgrade--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add grafana/ai-sdk --skill ai-sdk-parity-upgrade -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install grafana/ai-sdk ai-sdk-parity-upgrade --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/grafana/ai-sdk.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/.agents/skills/ai-sdk-parity-upgrade .gemini/skills/ai-sdk-parity-upgrade && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "ai-sdk-parity-upgrade" agent skill from https://github.com/grafana/ai-sdk/tree/main/.agents/skills/ai-sdk-parity-upgrade into .gemini/skills/ai-sdk-parity-upgrade/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "ai-sdk-parity-upgrade", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install grafana/ai-sdk ai-sdk-parity-upgradeInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add grafana/ai-sdk --skill ai-sdk-parity-upgrade -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/grafana/ai-sdk.git skills-src && mkdir -p .github/skills && cp -r skills-src/.agents/skills/ai-sdk-parity-upgrade .github/skills/ai-sdk-parity-upgrade && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "ai-sdk-parity-upgrade" agent skill from https://github.com/grafana/ai-sdk/tree/main/.agents/skills/ai-sdk-parity-upgrade into .github/skills/ai-sdk-parity-upgrade/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "ai-sdk-parity-upgrade", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add grafana/ai-sdk --skill ai-sdk-parity-upgrade -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install grafana/ai-sdk ai-sdk-parity-upgrade --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/grafana/ai-sdk.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/.agents/skills/ai-sdk-parity-upgrade .opencode/skills/ai-sdk-parity-upgrade && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "ai-sdk-parity-upgrade" agent skill from https://github.com/grafana/ai-sdk/tree/main/.agents/skills/ai-sdk-parity-upgrade into .opencode/skills/ai-sdk-parity-upgrade/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "ai-sdk-parity-upgrade", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
- Skill name
ai-sdk-parity-upgrade- GitHub stars
- 258
- Token cost
- ~1.9k tokens
- SKILL.md length
- 991 words
- Files
- 2
- Skills in repo
- 7
- Repo updated
- First seen
- Licence
- Apache-2.0
At a glance
Upgrade the pinned upstream AI SDK reference, assess the current Go implementation against it, and register parity work for independent implementation afterward.
- Works in 4 steps: Upgrade the pinned reference → Assess the current Go implementation… → Register the parity differences → …
- SKILL.md covers 1. Upgrade the pinned reference, 2. Assess the current Go…, 3. Register the parity… and 4. Implement parity work…
- Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md
What it does
AI SDK Parity Upgrade is an agent skill from grafana/ai-sdk, published by the product's own GitHub organization. Upgrade the pinned upstream AI SDK reference, assess the current Go implementation against it, and register parity work for independent implementation afterward.
Its SKILL.md is about 1.9k tokens, which your agent loads only when the skill is triggered. The skill folder holds 2 other files (for example `agents/openai.yaml`).
It works with Vercel AI SDK. The repository describes itself as: Grafana AI SDK for Go — streaming, tool-calling AI backends that speak fluent @ai-sdk/react. The licence is Apache-2.0.
Example prompts
- “/ai-sdk-parity-upgrade”
Workflow steps
4 steps, taken from the step headings in SKILL.md.
What it can do on your machine
Read from SKILL.md and the folder at commit bb0cacc. 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
AI SDK Parity Upgrade loads about 1.9k tokens when it runs. Until then it costs about 46 tokens; SKILL.md has 991 words of instructions outside code blocks.
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 grafana/ai-sdk at commit bb0cacc, republished under its Apache-2.0 licence (© grafana). 991 words, ~1,892 tokens.
Download SKILL.mdSave it as
.claude/skills/ai-sdk-parity-upgrade/SKILL.md (or your agent's skills folder). This skill also uses 1 other file; get the full folder from GitHub.- name
- ai-sdk-parity-upgrade
- description
- Upgrade the pinned upstream AI SDK reference, assess the current Go implementation against it, and register parity work for independent implementation afterward.
Pinned-Version Upgrade and Parity Assessment
Separate updating the upstream reference from matching all of its behavior:
- Pinned-version upgrade: update the reference, validate it, assess parity and register remaining differences. This produces one independently mergeable PR.
- Parity work: implement the registered behavioral work packages independently. This may produce several subsequent PRs without another upstream version bump.
The pinned version defines what we compare against; it does not claim exhaustive parity. A successful upgrade must leave a valid reference and an accountable assessment, not an empty parity backlog.
1. Upgrade the pinned reference
Use test/conformance/upstream.yaml for the starting versions and a fixed, coherent,
mature target for the destination. Update all consumer pins, the lockfile, generated
expectations and reviewed evidence together. Keep the starting references available
for explaining changed behavior. See the tooling reference
for selection, application and validation commands.
Run the tests and tooling against the target. Explain snapshot changes and resolve required-check failures. If the new reference introduces an incompatibility that prevents a supported integration from working, correct it here or wait to upgrade; registering a follow-up is not a substitute. This applies even when existing tests miss the incompatibility. Do not weaken comparisons to make the upgrade green.
These are upgrade blockers, not a requirement to implement every upstream feature before moving the pins. Other assessed differences can be registered for later work.
2. Assess the current Go implementation against the target
Perform a comprehensive comparison across the declared supported surfaces: core
orchestration, provider contracts/adapters, frontend behavior and Gateway. Include
harness limitations when evaluating the evidence. Use relevant PARITY.md entries
to identify current support, deviations and coverage.
Inspect exact upstream implementation and tests alongside the current Go paths. The release delta and changelogs guide investigation, but do not bound it: look for older gaps too. Trace requests, stream state, continuation, errors and interactions. Do not infer matching behavior from unchanged files or passing fixtures alone. Identify unsupported upstream product families separately, and state any area the assessment could not resolve rather than claiming it was covered.
| Finding | Decision | Evidence needed |
|---|---|---|
| Target breaks required checks or a supported integration | Resolve in the pinned-version upgrade or block it | Reproduction, source/test contract and regression proof |
| Go behavior differs without blocking the reference upgrade | Register a parity correction | Exact difference, impact and proposed acceptance tests |
| Go already matches | Retain it; register missing coverage if needed | Existing proof or a focused comparison |
| New capability within a supported area | Recommend a parity work package or explicit exclusion | Semantics, Go design implications and dependencies |
| Different mechanism, same observable behavior | Retain the Go adaptation | Behavioral and wire equivalence |
| Intentional observable difference or unsupported family | Confirm and document the disposition | Rationale and precise scope boundary |
| Behavior or impact is inconclusive | Investigate; keep the question visible | Missing source, reproduction or design decision |
Implementation and evidence are separate: a capability may work without sufficient proof, and a passing scenario may cover only part of it. Give new and existing gaps the same scrutiny. A known bug cannot disappear behind a coverage label, and an unresolved upgrade-blocking risk cannot be silently assigned to later work.
Show full SKILL.md (476 more words)Show less
3. Register the parity differences
Give each durable record one responsibility:
- GitHub issues own actionable deferred work. One
upstream-syncissue holds the behavior and impact, exact upstream reference, current Go difference, intended outcome, API decisions, dependencies, owner and acceptance evidence. Do not restate that content in repository documentation or mirror issue state. PARITY.mdowns stable coverage facts. Update it only when a surface's coverage classification, confidence source, supported boundary or accepted deviation changes. Never add a dated/versioned assessment section, issue catalog or upgrade-run ledger. A newly discovered actionable difference does not by itself require aPARITY.mdchange once an issue owns it.- The upgrade PR owns the run ledger. Record the target, corrections, validation and a compact exact list of created or reused issues. Summarize adaptations, exclusions and unresolved blockers there; promote only durable boundaries or accepted deviations to repository coverage metadata.
Search by behavior/provider across labeled and unlabeled issues and inspect open
and closed matches before creating anything. Reuse covered open work and flag
ambiguous overlaps rather than automatically duplicating it. Apply upstream-sync
to every new or reused parity issue, adding it when missing without replacing
other labels. Follow the issue registration rules
for search, issue contents and label selection, then link the issue from the
upgrade PR.
Registering work does not mean accepting a permanent deviation or approving a new API design. Obtain explicit scope/deviation decisions where needed. Reassess open packages against a later pinned reference instead of blindly carrying stale claims.
The upgrade PR is complete when the pins/evidence are consistent, required checks pass, the supported-surface assessment is accounted for and remaining differences are linked to work or an explicit disposition. A version bump alone is incomplete; a nonempty, honest parity inventory is not itself a failed upgrade.
4. Implement parity work independently
Select a registered package and confirm its contract against the current pinned reference. Refine the Go design and acceptance tests, then implement the complete behavior with regression proof. Prefer failing conformance cases with authentic inputs; otherwise use focused tests and state the remaining boundary coverage gap. Preserve fixture provenance. When work is completed, close or update its issue; change the coverage map only if the stable coverage status, evidence, support boundary or accepted deviation changed.
Work packages are behavioral units, not PRs. Related packages may share a PR; a package spanning published modules may require producer and consumer changes in separate PRs. Each PR must independently pass required checks, and the package is complete only when its full outcome and evidence are delivered. Go consumer pins change when they need published producer behavior, not automatically because the upstream reference changed.
Review in both directions: upstream → Go for missing semantics and edge cases, and Go → package for scope, design justification and regression proof. Use the parity-review skill for that review. Report package completion separately from completion of the earlier pinned-version upgrade.
© grafana, 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
SKILL.md and 1 other file in .agents/skills/ai-sdk-parity-upgrade of grafana/ai-sdk.
- SKILL.md
- agents/openai.yaml
Open the folder on GitHubat commit bb0cacc
Compare with similar skills
AI SDK Parity Upgrade 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.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| AI SDK Parity Upgrade this skillgrafana/ai-sdk | 258 | — | ~1.9k | Automated safety check: Pass | Apache-2.0 | |
| AI SDKvercel-labs/ai-facts | 168 | 20 repos | ~1.2k | Automated safety check: Pass | None | |
| AI Elementsxiaoiver/infinite-canvas-tutorial | 1.1k | 2 repos | ~1.9k | Automated safety check: Pass | MIT | |
| Mem0 Provider for Vercel AI SDKmem0ai/mem0 | 67k | — | ~2.3k | Automated safety check: Pass | Apache-2.0 | |
| AI Elements Chat Componentsmweinbach/agent-coworker | 156 | 5 repos | ~1.8k | Automated safety check: Pass | Custom licence | |
| Failproof AI SDK IntegrationFailproofAI/failproofai | 5.3k | — | ~6k | Automated safety check: Pass | Custom licence |
Similar skills
- Official
AI Elements
xiaoiver/infinite-canvas-tutorial
Create new AI chat interface components for the ai-elements library following established composable patterns, shadcn/ui integration, and Vercel AI SDK conventions.
Mem0 Provider for Vercel AI SDK
mem0ai/mem0
Adds persistent memory to Vercel AI SDK apps with the Mem0 provider, using a wrapped model or standalone retrieve and store utilities.
AI Elements Chat Components
mweinbach/agent-coworker
Adds AI Elements components such as conversations, messages, tool displays and prompt inputs to a Next.js project that uses shadcn/ui.
Failproof AI SDK Integration
FailproofAI/failproofai
Helps instrument a custom Python or TypeScript agent to record events for Failproof AI, verify what gets written, and run an evaluator worker that scores the runs.
Claude API
Kocoro-lab/Kocoro
Build apps with the Claude API or Anthropic SDK. An agent skill from Kocoro-lab/Kocoro.
More from grafana/ai-sdk
- Official
Openspec Archive Change
grafana/ai-sdk
Archive a completed OpenSpec change in the experimental workflow.
- Official
Openspec Explore
grafana/ai-sdk
Enter OpenSpec explore mode - a thinking partner for exploring ideas, investigating problems, and clarifying requirements in a project that uses OpenSpec.
- Official
Openspec Sync Specs
grafana/ai-sdk
Sync delta specs from an OpenSpec change to main specs. An agent skill from grafana/ai-sdk.
- Official
PR Description Writer
grafana/ai-sdk
Write or rewrite pull request descriptions from a branch diff, commit range, existing PR, issue, upstream reference, specification, bug report, or review context.
- Official
Openspec Verify Change
grafana/ai-sdk
Verify implementation matches OpenSpec change artifacts. An agent skill from grafana/ai-sdk.
- Official
AI SDK Parity Review
grafana/ai-sdk
Review an ai-sdk diff, feature, bug or package against its upstream parity contract.
Related
Works with
Questions about AI SDK Parity Upgrade
What does AI SDK Parity Upgrade do?
Upgrade the pinned upstream AI SDK reference, assess the current Go implementation against it, and register parity work for independent implementation afterward. AI SDK Parity Upgrade is an agent skill from grafana/ai-sdk, published by the product's own GitHub organization. Upgrade the pinned upstream AI SDK reference, assess the current Go implementation against it, and register parity work for independent implementation afterward.
How do I install AI SDK Parity Upgrade in Claude Code?
Run `npx skills add grafana/ai-sdk --skill ai-sdk-parity-upgrade -a claude-code`. Or copy the skill folder (.agents/skills/ai-sdk-parity-upgrade in grafana/ai-sdk) into .claude/skills/ai-sdk-parity-upgrade in your project. Claude Code loads it when a task matches its description.
How do I install AI SDK Parity Upgrade in Codex?
Run `npx skills add grafana/ai-sdk --skill ai-sdk-parity-upgrade -a codex`. Or copy the skill folder (.agents/skills/ai-sdk-parity-upgrade in grafana/ai-sdk) into .agents/skills/ai-sdk-parity-upgrade in your project. Codex loads it when a task matches its description.
Can I use AI SDK Parity Upgrade 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 grafana/ai-sdk --skill ai-sdk-parity-upgrade -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/ai-sdk-parity-upgrade, .gemini/skills/ai-sdk-parity-upgrade, .github/skills/ai-sdk-parity-upgrade and .opencode/skills/ai-sdk-parity-upgrade in your project.
What does AI SDK Parity Upgrade need to run?
SKILL.md names no scripts, command-line tools or credentials: AI SDK Parity Upgrade is instructions for the agent only.
Does AI SDK Parity Upgrade 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 AI SDK Parity Upgrade 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 AI SDK Parity Upgrade use?
AI SDK Parity Upgrade 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 AI SDK Parity Upgrade use?
About 1.9k tokens (SKILL.md is roughly 7.6k 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 AI SDK Parity Upgrade?
Skills that share tags, products or a category with AI SDK Parity Upgrade: AI SDK (vercel-labs/ai-facts, 168 stars), AI Elements (xiaoiver/infinite-canvas-tutorial, 1.1k stars), Mem0 Provider for Vercel AI SDK (mem0ai/mem0, 67k stars) and AI Elements Chat Components (mweinbach/agent-coworker, 156 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains AI SDK Parity Upgrade?
grafana (a GitHub organization, an official publisher) maintains it in grafana/ai-sdk, which has 258 GitHub stars. The repository holds 7 skills in this directory. The repository was last updated on October 9, 2026.
Source: grafana/ai-sdk on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.