foundry-iq — Claude Code skill: what it does, how to install
Sets up Foundry IQ knowledge bases in Azure, indexes local files or Blob storage, and routes retrieval to AI agents.
/plugin marketplace add microsoft/skills
/plugin install foundry-iq-skills@skillsWhat foundry-iq does
The skill handles the lifecycle of Foundry IQ knowledge bases on Azure. It indexes local files, Azure Blob, or ADLS storage into queryable knowledge sources and retrieves answers accompanied by citations. Before running any target query or plan, the assistant must read the specific procedure markdown file to follow strict operational boundaries.
Resource handling relies on an upfront prompt offering to reuse existing services, scan candidates, or create a new resource. Read queries execute without manual approvals. Writing or provisioning always demands an approved, unchanged plan first, while raw Search or Storage keys are strictly barred from the workflow.
Unsupported connectors or multi-source creation requests get diverted immediately to diagnostic routines before triggering cloud discovery. The tool does not manage other knowledge base vendors, repo-wide file search, or standalone Azure AI Search indexes outside of Foundry IQ.
What it helps with
- make local documents or Azure Blob storage searchable with citations
- attach an existing Foundry IQ knowledge base to an AI agent
- triage errors during multi-source knowledge base setup or connector provisioning
- clean up obsolete Foundry IQ resources using receipts
Example prompts for the agent
- Make our local policy documents searchable using Foundry IQ
- Connect the existing Azure knowledge base to our support agent
- Query the knowledge base about onboarding rules and include citations
How to install foundry-iq
1Plugin marketplace
/plugin marketplace add microsoft/skills/plugin install foundry-iq-skills@skillsThis installs the whole foundry-iq-skills plugin with all its skills
2Skills CLI (Vercel)
npx skills add microsoft/skills --skill foundry-iq -a claude-code3Manually: put the files in place
~/.claude/skills/foundry-iq/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/foundry-iq/
Manually: put the files in place
.agents/skills/foundry-iq/SKILL.mdFor all projects use ~/.agents/skills/
Manually: put the files in place
.cursor/skills/foundry-iq/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install microsoft/skills foundry-iq2Manually: put the files in place
.github/skills/foundry-iq/SKILL.mdPlatform CLI
gemini skills install https://github.com/microsoft/skills --path .github/plugins/foundry-iq-skills/skills/foundry-iqManually: put the files in place
.devin/skills/foundry-iq/SKILL.mdThe legacy .windsurf/skills/ path still works
Claude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about foundry-iq
How to install foundry-iq in Claude Code?
Run /plugin marketplace add microsoft/skills followed by /plugin install foundry-iq-skills@skills in your Claude Code session.
What does foundry-iq do?
It manages Azure Foundry IQ knowledge bases, helping create data sources, connect them to agents, and run citation-backed queries.
What tasks are outside the scope of foundry-iq?
It does not handle local repository code search, third-party knowledge base engines, or standard Azure AI Search setups that omit Foundry IQ.
Source text
SKILL.md · MIT
--- name: foundry-iq description: "Foundry IQ knowledge bases. WHEN: make local or Blob documents searchable; create/diagnose KBs; triage unsupported connectors or multi-source KB creation/reconfiguration (even with classic-search terms); connect existing KB to agents (including multi-source KBs); create/reuse a Search service; retrieve from an existing knowledge base with citations; no brand words needed. Read its procedure before query/target questions. NOT: other KB providers, repository-file search, classic Azure AI Search index/query/app work with no Foundry IQ KB or connector request, generic agent creation." license: MIT compatibility: Azure metadata: author: Microsoft version: "0.2.1" --- # Foundry IQ Read one procedure before questions/actions; invocation is not a read. Read the owning procedure before blocked/unsupported responses too. Before mutation plans/approval, successfully read its required pre-action references, selected branches only. Do not reload successful reads. On failure, try only permitted bounded exact-path reads with any supported reader; never bypass restrictions or search broadly. If still unavailable: `blocked: reference-unavailable`, name missing references and inability to plan. Never invent requirements/plans. Failures first. Cleanup and receipt-backed execution go directly to their lifecycle/producer owner; do not reopen Search intake, provisioning or hardening. Before expensive service discovery, reuse supplied resource/intent; otherwise ask one early **USE EXISTING / FIND CANDIDATES / CREATE NEW** choice. Only FIND enumerates; supplied identity uses exact/minimum scoped resolution. CREATE checks its proposed name, not existing-service inventories. Preserve these answers across source/KB/CU/model handoffs; selection is not write approval. Searchable docs: KB + validated retrieval. Confirm intent once; KS-only must be explicit. Child success is not KB completion. Unclear/compound/mode/completion: [read](references/intent-routing.md). Route by requested operation, not existing KB source count. Connecting an existing KB with two or more sources, without changing the KB, uses Connect. Read-only operations use their owning procedure and actual helper constraints; this does not add retrieval modes or supported source kinds. Explicit unsupported provisioning or multi-source KB creation/reconfiguration uses Diagnose and stops before discovery, even when connecting an agent is also requested. This routing takes precedence over classic-search terms (e.g. hybrid search) in compound requests. Never end in a plain refusal: Diagnose gives the boundary, docs pointer and agent handoff; writes still need an approved concrete plan. Do not silently execute only the supported part of a compound request. If existing-KB connection versus KB creation/reconfiguration is unclear, read intent-routing and clarify that scope before Azure discovery. Never infer KB mutation. |Outcome|Read| |---|---| |Cleanup|[Plan cleanup](lifecycle/cleanup.md)| | Failure/drift | [Diagnose](troubleshooting/diagnose.md) | | Unsupported connector provisioning / multi-source KB creation or reconfiguration | [Diagnose](troubleshooting/diagnose.md) | | Connect | [Connect](agents/connect.md) | | Read KB | [Query](knowledge-bases/retrieve.md) | | Search only | [Search](search-services/create.md) | | File KS only | [File](knowledge-sources/create-file.md) | | Blob/ADLS KS only | [Blob](knowledge-sources/create-azure-blob.md) | | Searchable/KB | [KB](knowledge-bases/create.md) | Generic agents: `microsoft-foundry`. Reads: no approval; approve unchanged plans before writes. Hide hashes. No Search/Storage keys, scope widening, guessed identity/boundaries, drift repair or joint cleanup/creation approval. Acceptance != success.
© Microsoft · MIT · microsoft/skills