Platform Awareness
Owns the deferred-tool discipline, the VS Code Copilot platform-change tracking, and the SKILL.md description discipline that platform surfaces expose. This skill's description front-loads discovery vocabulary (GitHub, Azure, Fabric, Microsoft docs, browser, notebook, mermaid, Bicep, Figma, Microsoft Graph, MCP) so it fires reliably on any request that touches those surfaces.
Deferred Tools — Search Before Calling
Many tools are deferred (lazy-loaded) as of VS Code 1.118+. They appear in availableDeferredTools but cannot be called directly. Load via tool_search first with a natural-language capability description.
Rules
- Search before calling. Calling a deferred tool without loading via
tool_search fails silently.
- Search once per tool. After load, the tool stays available for the session.
- Use broad queries. One broad search beats multiple narrow ones.
- No results means unavailable. Don't retry with synonyms.
External Ingest (VS Code 1.119+)
In remote or virtual-filesystem workspaces (GitHub.dev, VS Code Remote, Codespaces), the editor provides codebase context automatically. semantic_search and file operations work transparently — no agent action needed.
Common Deferred Tool Categories
Reach for these search queries when a specific capability is needed:
| Category |
Example tools |
Search query |
| GitHub operations |
issues, PRs, repos, code search, branches, tags |
github |
| Azure MCP |
Storage, KeyVault, Cosmos, SQL, AKS, App Service (48+ services) |
azure or the specific service name |
| Microsoft Fabric |
Eventstream, Kusto, OneLake, items |
fabric or onelake |
| Microsoft docs |
docs search, code samples, full-page fetch |
microsoft docs |
| Browser automation |
click, navigate, screenshot, fill form, evaluate JS |
browser or playwright |
| Notebook operations |
run cell, edit notebook, read output |
notebook |
| Mermaid rendering |
preview, validate diagrams |
mermaid |
| Bicep / ARM |
best practices, schema, diagnostics, format |
bicep |
| Figma |
design context, code connect, screenshots |
figma |
| Microsoft Graph |
get, list, suggest queries |
microsoft graph or entra |
Do not hardcode tool names from availableDeferredTools without loading them via tool_search. The list is informational; actual availability requires the search call. If the topic is not obviously in the categories above, broaden the search query empirically — start with one word, iterate.
VS Code 1.122–1.128 conveniences
| Release |
Capability |
What it changes for me |
| 1.122 |
/models slash command |
Opens the model picker from chat input. Useful when the user asks to switch models mid-task without leaving chat. |
| 1.122 |
BYOK air-gapped |
Bring Your Own Key models work without GitHub authentication. Heirs in regulated/enterprise contexts can run Copilot Chat fully offline; the BYOK token counter (introduced 1.120) keeps working. |
| 1.122 |
Local agent host default-on (Insiders only) |
Watchpoint: when this reaches Stable, deferred-tool resolution may shift. No action until first observed behavior change. |
| 1.123 |
Session sync + /chronicle |
Chat sessions auto-sync to the GitHub account (gated by chat.sessionSync.enabled, org-managed). The /chronicle slash command queries that history. The brain's own chronicle skill is local-only; the platform feature is an adjacent capability, not a replacement. |
| 1.123 |
Sandbox network-retry |
When a local-agent terminal command needs an unallowed domain, VS Code auto-retries inside the sandbox with unrestricted network before falling back to unsandboxed (chat.agent.sandbox.retryWithAllowNetworkRequests). Reduces spurious failures on git fetch / npm install. |
| 1.124 |
Autopilot enabled by default |
Autopilot Preview is now on by default; chat.permissions.default controls the per-workspace level. ACT's heir-workspace baseline pins default as the deliberate opt-out — see heir-workspace-settings-baseline.json. |
| 1.124 |
Advanced Autopilot (opt-in) |
chat.autopilot.advanced.enabled uses a utility model to judge when a task is truly done. Capped at 3 iterations. Off by default; opt-in only. |
| 1.124 |
Enterprise Copilot plugin policies |
chat.plugins.enabledPlugins / chat.plugins.extraMarketplaces / chat.plugins.strictMarketplaces let org admins allowlist plugin IDs and marketplaces. Heirs in regulated orgs may see Mall installs silently blocked or marketplaces tagged as policy-managed; surface a clear message rather than retrying when an install refuses. |
| 1.125 |
extensions.autoUpdate simplified to on / off |
Old values (true / false / onlyEnabledExtensions / delayed) migrate automatically. A project that pins settings should pin the canonical on shape from this release forward; pre-1.125 heirs that have the old value still work via migration. |
| 1.125 |
extensions.autoUpdateDelay configurable |
The 2-hour supply-chain delay introduced in 1.123 is now a configurable hour count. Alex ACT pins no value here, so heirs ride the platform default, but the setting exists if a heir or org wants tighter or looser update windows. |
| 1.125 |
Forwarded-port URL rewriting for agents |
When an agent in a remote workspace requests a port that has been forwarded, VS Code rewrites the URL and notifies the agent of the change. Reduces spurious browser-tool failures in Codespaces / Remote-SSH heir setups. No action needed; surface in fetch/browser diagnostics if a heir reports a port mismatch. |
| 1.125 |
Native MDM delivery for managed Copilot settings |
On Windows/macOS, org admins can deliver Copilot settings via MDM channels in addition to the account-based enterprise file. Settings delivered via MDM appear as policy-enforced and cannot be overridden locally — heirs in MDM-managed orgs may see baseline welcome-baseline.json keys ignored if the MDM policy disagrees. |
| 1.126 |
Edit mode deprecated → Agent mode |
chat.editMode.hidden also removed. Reinforces welcome-baseline chat.agent.enabled: true pin — user policy and platform default converged. Heirs with agent mode policy-disabled still see legacy Edit mode. |
| 1.127 |
macOS/Linux terminal sandboxing |
Agent-invoked terminal commands run with network blocked + FS restricted; agent only prompts on elevation. Substantially reduces approval prompts on non-Windows heirs. Toggle in Permissions dropdown. Windows unaffected — Backtick Hazard + Output Capture rules still apply. |
| 1.127 |
/troubleshoot on agent host sessions |
Adjacent to the brain's ACT: Diagnose Fetch pattern — platform-side diagnostic for agent behavior questions (custom instructions ignored, slow responses); not a replacement for our fetch diagnostics. |
| 1.127 |
Browser tools GA (workbench.browser.enableChatTools) |
Agent can open pages, screenshot, click through to validate its own work. Org-managed setting. Enterprise policies BrowserChatTools + ChatAgentNetworkFilter (agent domain allow/deny lists) may block Mall/GitHub fetches for regulated heirs — surface a clear message rather than retrying when a fetch refuses. |
| 1.127 |
File-based managed Copilot settings |
Extends 1.125 MDM channel: managed-settings.json at %ProgramFiles%\GitHubCopilot\ (Win) / /Library/Application Support/GitHubCopilot/ (mac) / /etc/github-copilot/ (Linux). Same policy-lockout semantics as the MDM row — baseline keys may be ignored in policy-managed environments. |
| 1.127 |
Built-in Ollama provider deprecated |
Official Ollama VS Code extension replaces it. Heirs using BYOK Ollama should install the extension and remove the built-in provider. No brain action; documents why welcome-baseline.json doesn't pin an Ollama provider. |
| 1.128 |
BYOK Claude via own Anthropic key |
Heirs can run the Claude agent through their own Anthropic API credentials instead of consuming GitHub Copilot quota. Relevant wherever a project weighs the credit economy of model choice. No baseline setting; user-opt-in via Claude harness auth. |
| 1.128 |
Custom endpoint model options for BYOK |
Enables BYOK against strict-schema providers (Moonshot, Kimi, etc.) that reject non-standard params. temperature and provider-specific options now configurable. Unblocks heirs who reported provider-rejection errors before this. |
| 1.128 |
Claude agent → integrated browser DOM/tools |
Feature parity with the Copilot agent's browser tools GA (1.127). Same BrowserChatTools + ChatAgentNetworkFilter enterprise policies apply. |
Skill Picker — Never Strip the Description
The SKILL.md description field has three consumers and the slash-picker tooltip is the least important of them:
- Agent skill discovery (primary) — every session loads SKILL.md descriptions into the
<skills> block; this is how the parent agent decides whether to invoke the skill
- Brain QA enforcement — where a project ships a brain-QA script, it typically hard-fails on a missing or empty description
- Chat picker tooltip — the surface visible to humans
Keep Agent Skills enabled, but disable the experimental generic resolver.
VS Code 1.131 advertises plugin-contributed skills that its generic skill
tool cannot resolve. Until
microsoft/vscode#314772
closes and a plugin-skill probe passes, use this user-level shape:
"chat.useAgentSkills": true,
"github.copilot.chat.skillTool.enabled": false
This restores file-based loading and namespaced command fallbacks. Never strip
the skill description as a workaround; descriptions remain the discovery
signal.
Related
- browser-tools skill — Playwright / browser automation specifics (uses the deferred-tool discipline this skill owns)
skill-review / skill-creator (Alex ACT Steward, not shipped in Core) — the frontmatter description discipline this skill guards against being stripped
Would Revise If
Revise if the categories table goes stale (VS Code adds new deferred tool families not listed), if the search-query patterns produce zero results for tools that availableDeferredTools lists as present, if the platform-changes table drifts more than 2 VS Code releases behind current, or if VS Code renames github.copilot.chat.skillTool.enabled.
Skill resolver section falsifier: revise by 2026-11-03 or sooner if VS Code renames or removes github.copilot.chat.skillTool.enabled, closes microsoft/vscode#314772, or an enabled plugin skill succeeds through the generic skill tool in a clean Agent mode session. First observed contradiction wins.
1---2name: platform-awareness3description: VS Code Copilot platform changes affecting how tools are used: deferred-tool categories with example search queries (GitHub, Azure MCP, Fabric, Microsoft docs, browser, notebook, mermaid, Bicep, Figma, Microsoft Graph), the VS Code 1.117-1.128 platform-change table (models, BYOK, sandboxing, browser tools GA, Ollama, Claude BYOK), and skill-picker surfacing behavior (github.copilot.chat.skillTool.enabled). Use when working with deferred tools, MCP servers, GitHub APIs, notebooks, browser automation, when asking about VS Code Copilot Chat platform features, or when a tool_search call needs a category suggestion.4---56# Platform Awareness78Owns the deferred-tool discipline, the VS Code Copilot platform-change tracking, and the SKILL.md description discipline that platform surfaces expose. This skill's description front-loads discovery vocabulary (GitHub, Azure, Fabric, Microsoft docs, browser, notebook, mermaid, Bicep, Figma, Microsoft Graph, MCP) so it fires reliably on any request that touches those surfaces.910## Deferred Tools — Search Before Calling1112Many tools are **deferred** (lazy-loaded) as of VS Code 1.118+. They appear in `availableDeferredTools` but cannot be called directly. Load via `tool_search` first with a natural-language capability description.1314### Rules15161. **Search before calling.** Calling a deferred tool without loading via `tool_search` fails silently.172. **Search once per tool.** After load, the tool stays available for the session.183. **Use broad queries.** One broad search beats multiple narrow ones.194. **No results means unavailable.** Don't retry with synonyms.2021## External Ingest (VS Code 1.119+)2223In remote or virtual-filesystem workspaces (GitHub.dev, VS Code Remote, Codespaces), the editor provides codebase context automatically. `semantic_search` and file operations work transparently — no agent action needed.2425## Common Deferred Tool Categories2627Reach for these search queries when a specific capability is needed:2829| Category | Example tools | Search query |30| --- | --- | --- |31| GitHub operations | issues, PRs, repos, code search, branches, tags | `github` |32| Azure MCP | Storage, KeyVault, Cosmos, SQL, AKS, App Service (48+ services) | `azure` or the specific service name |33| Microsoft Fabric | Eventstream, Kusto, OneLake, items | `fabric` or `onelake` |34| Microsoft docs | docs search, code samples, full-page fetch | `microsoft docs` |35| Browser automation | click, navigate, screenshot, fill form, evaluate JS | `browser` or `playwright` |36| Notebook operations | run cell, edit notebook, read output | `notebook` |37| Mermaid rendering | preview, validate diagrams | `mermaid` |38| Bicep / ARM | best practices, schema, diagnostics, format | `bicep` |39| Figma | design context, code connect, screenshots | `figma` |40| Microsoft Graph | get, list, suggest queries | `microsoft graph` or `entra` |4142Do not hardcode tool names from `availableDeferredTools` without loading them via `tool_search`. The list is informational; actual availability requires the search call. If the topic is not obviously in the categories above, broaden the search query empirically — start with one word, iterate.4344## VS Code 1.122–1.128 conveniences4546| Release | Capability | What it changes for me |47| ------- | ------------------------------------------------------ | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |48| 1.122 | `/models` slash command | Opens the model picker from chat input. Useful when the user asks to switch models mid-task without leaving chat. |49| 1.122 | BYOK air-gapped | Bring Your Own Key models work without GitHub authentication. Heirs in regulated/enterprise contexts can run Copilot Chat fully offline; the BYOK token counter (introduced 1.120) keeps working. |50| 1.122 | Local agent host default-on (Insiders only) | Watchpoint: when this reaches Stable, deferred-tool resolution may shift. No action until first observed behavior change. |51| 1.123 | Session sync + `/chronicle` | Chat sessions auto-sync to the GitHub account (gated by `chat.sessionSync.enabled`, org-managed). The `/chronicle` slash command queries that history. The brain's own `chronicle` skill is local-only; the platform feature is an adjacent capability, not a replacement. |52| 1.123 | Sandbox network-retry | When a local-agent terminal command needs an unallowed domain, VS Code auto-retries inside the sandbox with unrestricted network before falling back to unsandboxed (`chat.agent.sandbox.retryWithAllowNetworkRequests`). Reduces spurious failures on `git fetch` / `npm install`. |53| 1.124 | Autopilot enabled by default | Autopilot Preview is now on by default; `chat.permissions.default` controls the per-workspace level. ACT's heir-workspace baseline pins `default` as the deliberate opt-out — see `heir-workspace-settings-baseline.json`. |54| 1.124 | Advanced Autopilot (opt-in) | `chat.autopilot.advanced.enabled` uses a utility model to judge when a task is truly done. Capped at 3 iterations. Off by default; opt-in only. |55| 1.124 | Enterprise Copilot plugin policies | `chat.plugins.enabledPlugins` / `chat.plugins.extraMarketplaces` / `chat.plugins.strictMarketplaces` let org admins allowlist plugin IDs and marketplaces. Heirs in regulated orgs may see Mall installs silently blocked or marketplaces tagged as policy-managed; surface a clear message rather than retrying when an install refuses. |56| 1.125 | `extensions.autoUpdate` simplified to `on` / `off` | Old values (`true` / `false` / `onlyEnabledExtensions` / `delayed`) migrate automatically. A project that pins settings should pin the canonical `on` shape from this release forward; pre-1.125 heirs that have the old value still work via migration. |57| 1.125 | `extensions.autoUpdateDelay` configurable | The 2-hour supply-chain delay introduced in 1.123 is now a configurable hour count. Alex ACT pins no value here, so heirs ride the platform default, but the setting exists if a heir or org wants tighter or looser update windows. |58| 1.125 | Forwarded-port URL rewriting for agents | When an agent in a remote workspace requests a port that has been forwarded, VS Code rewrites the URL and notifies the agent of the change. Reduces spurious browser-tool failures in Codespaces / Remote-SSH heir setups. No action needed; surface in fetch/browser diagnostics if a heir reports a port mismatch. |59| 1.125 | Native MDM delivery for managed Copilot settings | On Windows/macOS, org admins can deliver Copilot settings via MDM channels in addition to the account-based enterprise file. Settings delivered via MDM appear as policy-enforced and cannot be overridden locally — heirs in MDM-managed orgs may see baseline `welcome-baseline.json` keys ignored if the MDM policy disagrees. |60| 1.126 | Edit mode deprecated → Agent mode | `chat.editMode.hidden` also removed. Reinforces welcome-baseline `chat.agent.enabled: true` pin — user policy and platform default converged. Heirs with agent mode policy-disabled still see legacy Edit mode. |61| 1.127 | macOS/Linux terminal sandboxing | Agent-invoked terminal commands run with network blocked + FS restricted; agent only prompts on elevation. Substantially reduces approval prompts on non-Windows heirs. Toggle in Permissions dropdown. Windows unaffected — Backtick Hazard + Output Capture rules still apply. |62| 1.127 | `/troubleshoot` on agent host sessions | Adjacent to the brain's `ACT: Diagnose Fetch` pattern — platform-side diagnostic for agent behavior questions (custom instructions ignored, slow responses); not a replacement for our fetch diagnostics. |63| 1.127 | Browser tools GA (`workbench.browser.enableChatTools`) | Agent can open pages, screenshot, click through to validate its own work. Org-managed setting. Enterprise policies `BrowserChatTools` + `ChatAgentNetworkFilter` (agent domain allow/deny lists) may block Mall/GitHub fetches for regulated heirs — surface a clear message rather than retrying when a fetch refuses. |64| 1.127 | File-based managed Copilot settings | Extends 1.125 MDM channel: `managed-settings.json` at `%ProgramFiles%\GitHubCopilot\` (Win) / `/Library/Application Support/GitHubCopilot/` (mac) / `/etc/github-copilot/` (Linux). Same policy-lockout semantics as the MDM row — baseline keys may be ignored in policy-managed environments. |65| 1.127 | Built-in Ollama provider deprecated | Official Ollama VS Code extension replaces it. Heirs using BYOK Ollama should install the extension and remove the built-in provider. No brain action; documents why `welcome-baseline.json` doesn't pin an Ollama provider. |66| 1.128 | BYOK Claude via own Anthropic key | Heirs can run the Claude agent through their own Anthropic API credentials instead of consuming GitHub Copilot quota. Relevant wherever a project weighs the credit economy of model choice. No baseline setting; user-opt-in via Claude harness auth. |67| 1.128 | Custom endpoint model options for BYOK | Enables BYOK against strict-schema providers (Moonshot, Kimi, etc.) that reject non-standard params. `temperature` and provider-specific options now configurable. Unblocks heirs who reported provider-rejection errors before this. |68| 1.128 | Claude agent → integrated browser DOM/tools | Feature parity with the Copilot agent's browser tools GA (1.127). Same `BrowserChatTools` + `ChatAgentNetworkFilter` enterprise policies apply. |6970## Skill Picker — Never Strip the Description7172The SKILL.md `description` field has three consumers and the slash-picker tooltip is the least important of them:73741. **Agent skill discovery (primary)** — every session loads SKILL.md descriptions into the `<skills>` block; this is how the parent agent decides whether to invoke the skill752. **Brain QA enforcement** — where a project ships a brain-QA script, it typically hard-fails on a missing or empty description763. **Chat picker tooltip** — the surface visible to humans7778**Keep Agent Skills enabled, but disable the experimental generic resolver.**79VS Code 1.131 advertises plugin-contributed skills that its generic `skill`80tool cannot resolve. Until81[microsoft/vscode#314772](https://github.com/microsoft/vscode/issues/314772)82closes and a plugin-skill probe passes, use this user-level shape:8384```jsonc85"chat.useAgentSkills": true,86"github.copilot.chat.skillTool.enabled": false87```8889This restores file-based loading and namespaced command fallbacks. Never strip90the skill description as a workaround; descriptions remain the discovery91signal.9293## Related9495- [browser-tools skill](../browser-tools/SKILL.md) — Playwright / browser automation specifics (uses the deferred-tool discipline this skill owns)96- `skill-review` / `skill-creator` (Alex ACT Steward, not shipped in Core) — the frontmatter description discipline this skill guards against being stripped9798## Would Revise If99100Revise if the categories table goes stale (VS Code adds new deferred tool families not listed), if the search-query patterns produce zero results for tools that `availableDeferredTools` lists as present, if the platform-changes table drifts more than 2 VS Code releases behind current, or if VS Code renames `github.copilot.chat.skillTool.enabled`.101102**Skill resolver section falsifier**: revise by 2026-11-03 or sooner if VS Code renames or removes `github.copilot.chat.skillTool.enabled`, closes microsoft/vscode#314772, or an enabled plugin skill succeeds through the generic `skill` tool in a clean Agent mode session. First observed contradiction wins.
fabioc-aloha