feat: improve the hosted agent getting started with coding by anchenyi · Pull Request #2610 · microsoft/GitHub-Copilot-for-Azure

anchenyi · 2026-06-09T09:22:05Z

Description

Continue development based on #2486 and #2435

Checklist

Tests pass locally (cd tests && npm test)
Title has one of the prefixes: fix:, feat:, feature:, chore:, misc:, test:, eval:
If modifying skill descriptions: verified routing correctness with integration tests (In tests/, npm run test:integration -- <skill> or npm run test:vally -- --skill <skill>)

Related Issues

…eploy workflows - Streamline create-hosted.md and deploy.md with clearer step-by-step flows - Add reference docs for azd-ai-cli, local-run, and tools - Add environment verification scripts (PowerShell and Bash) - Update direct-code-deployment reference with consistent CPU/memory defaults - Update SKILL.md description for better routing Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

The previous defaults of 0.5 CPU / 1Gi often cause session_not_ready timeouts during cold start. Update all prose and JSON examples to use 1 CPU / 2Gi as the recommended default. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

…testing

* fix: restore eval * fix: comment * fix: use azd eval

Copilot

Pull request overview

This PR continues the microsoft-foundry skill refactor to make the “hosted agent getting started” flow primarily azd-first (scaffold → local run → deploy → invoke), while removing older/deprecated direct-code REST guidance and adding helper scripts/reference docs to reduce setup friction.

Changes:

Updates the microsoft-foundry top-level routing description and sub-skill summaries to emphasize azd-based hosted-agent workflows.
Rewrites the hosted-agent create and deploy docs around azd ai agent init/run/deploy/invoke, and adds new reference docs (azd CLI mental model, local run, tools/toolboxes).
Adds environment verification + project-id resolution helper scripts (bash/PowerShell) and removes the old direct-code deployment reference.

Reviewed changes

Copilot reviewed 13 out of 14 changed files in this pull request and generated 7 comments.

Show a summary per file

File	Description
plugin/skills/microsoft-foundry/SKILL.md	Updates skill description and sub-skill “when to use” blurbs to reflect azd-centric hosted-agent lifecycle.
plugin/skills/microsoft-foundry/project/create/create-foundry-project.md	Adds auto-generated project naming guidance during project creation flow.
plugin/skills/microsoft-foundry/foundry-agent/invoke/invoke.md	Adds an azd-first “fast smoke test” path and expands troubleshooting entries for azd invoke.
plugin/skills/microsoft-foundry/foundry-agent/deploy/references/direct-code-deployment.md	Removes deprecated direct-code REST deployment reference.
plugin/skills/microsoft-foundry/foundry-agent/deploy/deploy.md	Major rewrite of deploy workflow to focus on `azd provision/deploy` and async eval init.
plugin/skills/microsoft-foundry/foundry-agent/create/scripts/verify-environment.sh	Adds a bash script to validate azd + extensions + auth + basic project/agent status.
plugin/skills/microsoft-foundry/foundry-agent/create/scripts/verify-environment.ps1	Adds a PowerShell equivalent environment verification script.
plugin/skills/microsoft-foundry/foundry-agent/create/scripts/resolve-project-id.sh	Adds a bash helper to resolve Foundry project ARM ID from a project endpoint via Azure CLI.
plugin/skills/microsoft-foundry/foundry-agent/create/scripts/resolve-project-id.ps1	Adds a PowerShell helper to resolve Foundry project ARM ID from a project endpoint via Azure CLI.
plugin/skills/microsoft-foundry/foundry-agent/create/references/tools.md	Adds toolbox/connection recipes and MCP wiring guidance for hosted agents.
plugin/skills/microsoft-foundry/foundry-agent/create/references/local-run.md	Adds detailed guidance for running hosted agents locally with `azd ai agent run`.
plugin/skills/microsoft-foundry/foundry-agent/create/references/azd-ai-cli.md	Adds an “azd ai CLI” mental model doc describing file roles and command surface.
plugin/skills/microsoft-foundry/foundry-agent/create/create-hosted.md	Rewrites hosted-agent creation workflow around azd samples + brownfield init + local iteration.
.gitignore	Ignores `tmp/` at repo root.

fix(microsoft-foundry): clarify ENABLE_HOSTED_AGENTS vs ENABLE_CAPABILITY_HOST for azd ai agent

refactor(microsoft-foundry): teach correct azd Golden Path for model deployments

…install and provision run simultaneously

feat: install package in parallel and make remote invocation runs just one time

XOEEst · 2026-06-11T07:25:52Z

Looks good. Can you provide a conversation history to show the no-wait experience of eval init? E.g. will it prompt for check status after deploy completes?

And the change of eval init -> eval generate has been released in latest azd extension:
azd ext install azure.ai.agents --version 0.1.39-preview

swatDong · 2026-06-11T08:03:56Z

Looks good. Can you provide a conversation history to show the no-wait experience of eval init? E.g. will it prompt for check status after deploy completes?

And also please wait for eval init -> eval generate to save future migration efforts. :)

Per test, agent auto-asks for user selection on eval:

Then call eval init --no-wait to trigger the generation:

And finally in summary mentioned eval suite generation in progress and eval run as next step:

fix: improve parallel install and local run stability

refactor: remove parallel install

chore: improve local run stability

* feat: golden path subskill * feat: quick start subskill * feat: quick start subskill * feat: quick start subskill

XOEEst · 2026-06-13T09:42:17Z

Looks good. Can you provide a conversation history to show the no-wait experience of eval init? E.g. will it prompt for check status after deploy completes?
And also please wait for eval init -> eval generate to save future migration efforts. :)

Per test, agent auto-asks for user selection on eval:

Then call eval init --no-wait to trigger the generation:

And finally in summary mentioned eval suite generation in progress and eval run as next step:

In the last snapshot, will eval run wait for eval suite generation to complete? I guess not, so shall we add a next step to wait for eval suite generation to finish?

swatDong · 2026-06-15T02:24:14Z

Looks good. Can you provide a conversation history to show the no-wait experience of eval init? E.g. will it prompt for check status after deploy completes?
And also please wait for eval init -> eval generate to save future migration efforts. :)

Per test, agent auto-asks for user selection on eval:
Then call eval init --no-wait to trigger the generation:
And finally in summary mentioned eval suite generation in progress and eval run as next step:

In the last snapshot, will eval run wait for eval suite generation to complete? I guess not, so shall we add a next step to wait for eval suite generation to finish?

Tried eval generate --no-wait and eval run, eval run does wait for eval suite generation to complete, so IMO the current eval run next step is enough.

chore: update tests

Copilot

Pull request overview

Copilot reviewed 38 out of 39 changed files in this pull request and generated 3 comments.

zhenjiao-ms and others added 11 commits June 2, 2026 16:23

feat: improve local run, deploy, and invoke. Add temp eval infra for …

cd5c2a9

…testing

refactor: deploy sub skill and minors

234d0f3

Merge branch 'main' into dev/improve-getting-started

ea25cbe

feat: swtich to acr as default, improve wording

c33b248

feat: direct code as default

b63e4d1

chore: align env with agent.yaml

c1de7d6

fix: reduce golden time (microsoft#2593)

5a0fd6a

fix: venv and arm id

d64e2fa

fix: restore eval (microsoft#2606)

af0f2a8

* fix: restore eval * fix: comment * fix: use azd eval

Copilot AI review requested due to automatic review settings June 9, 2026 09:22

anchenyi requested review from RickWinter, XOEEst, XiaofuHuang, ankitbko, jugonzales, tendau and vebudumu as code owners June 9, 2026 09:22

Copilot started reviewing on behalf of anchenyi June 9, 2026 09:22 View session

Copilot AI reviewed Jun 9, 2026

View reviewed changes

dooriya and others added 9 commits June 9, 2026 18:04

fix: capability host setup for hosted agent

cbfe22a

fix: minor tweak

c8bc22d

Merge pull request #2 from anchenyi/dol/fix-capability-host

f40cad2

fix(microsoft-foundry): clarify ENABLE_HOSTED_AGENTS vs ENABLE_CAPABILITY_HOST for azd ai agent

refactor(microsoft-foundry): teach azd Golden Path for model deployments

e86c980

fix: tweak wording

467ea37

Merge pull request #3 from anchenyi/dol/improve-model-deploy

606697a

refactor(microsoft-foundry): teach correct azd Golden Path for model deployments

feat: fix model deployment name, let agent only invoke once, make uv …

3b7b6b1

…install and provision run simultaneously

chore: improve uv install

7de2714

chore: clean

a5e10c5

anchenyi and others added 3 commits June 10, 2026 15:43

chore: improve async install

9e7c188

Merge pull request #5 from anchenyi/anchenyi/async-install

db9c43e

feat: install package in parallel and make remote invocation runs just one time

fix: improve install and local run command

6813a53

chore: simplify local run prompt, add local clean up after local testing

4dec589

Merge pull request #6 from anchenyi/anchenyi/improve-install

bcfb162

fix: improve parallel install and local run stability

github-actions Bot mentioned this pull request Jun 11, 2026

[repo-status] Weekly Repo Status — June 11, 2026 #2626

Closed

anchenyi and others added 8 commits June 12, 2026 10:22

refactor: remove parallel install

ebd6e53

Merge pull request #7 from anchenyi/anchenyi/remove-parallel-install

451e2c5

refactor: remove parallel install

fix: latest eval generate (#8)

dc64e91

chore: improve local run

4cb6f51

chore: improve local run

3336e99

chore: update

7d8c6d8

Merge pull request #9 from anchenyi/anchenyi/improve-local-run

d38c94f

chore: improve local run stability

feat: self-contained golden path sub skill (#10)

2ea78c8

* feat: golden path subskill * feat: quick start subskill * feat: quick start subskill * feat: quick start subskill

github-actions Bot mentioned this pull request Jun 12, 2026

[repo-status] Weekly Repo Status — June 12, 2026 #2631

Closed

swatDong and others added 3 commits June 15, 2026 11:27

fix: local run (#11)

b68ad65

chore: update tests

6e2d9d0

Merge pull request #12 from anchenyi/anchenyi/add-tests

b37ed48

chore: update tests

anchenyi requested a review from Copilot June 15, 2026 06:53

Copilot started reviewing on behalf of anchenyi June 15, 2026 06:53 View session

XiaofuHuang approved these changes Jun 15, 2026

View reviewed changes

Copilot AI reviewed Jun 15, 2026

View reviewed changes

Comment thread plugin/skills/microsoft-foundry/foundry-agent/create/create-hosted.md

Comment thread plugin/skills/microsoft-foundry/foundry-agent/create/references/azd-ai-cli.md

Comment thread plugin/skills/microsoft-foundry/foundry-agent/invoke/invoke.md

anchenyi changed the title ~~[WIP] feat: improve the hosted agent getting started with coding~~ feat: improve the hosted agent getting started with coding Jun 15, 2026

anchenyi merged commit 5812e44 into microsoft:main Jun 15, 2026
11 of 12 checks passed

This was referenced Jun 18, 2026

[repo-status] Weekly Repo Status — June 13–18, 2026 #2668

Closed

[repo-status] Weekly Repo Status — June 13–18, 2026 #2670

Closed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

feat: improve the hosted agent getting started with coding#2610

feat: improve the hosted agent getting started with coding#2610
anchenyi merged 38 commits into
microsoft:mainfrom
anchenyi:improve-getting-started

anchenyi commented Jun 9, 2026

Uh oh!

Copilot AI left a comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

XOEEst commented Jun 11, 2026 •

edited

Loading

Uh oh!

swatDong commented Jun 11, 2026

Uh oh!

XOEEst commented Jun 13, 2026

Uh oh!

swatDong commented Jun 15, 2026

Uh oh!

Copilot AI left a comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

7 participants

Uh oh!

Conversation

anchenyi commented Jun 9, 2026

Description

Checklist

Related Issues

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

XOEEst commented Jun 11, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

swatDong commented Jun 11, 2026

Uh oh!

XOEEst commented Jun 13, 2026

Uh oh!

swatDong commented Jun 15, 2026

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

7 participants

XOEEst commented Jun 11, 2026 •

edited

Loading