Tessl • The management layer for AI agent skills

SkillPublic

policy-and-managed-settings

Use whenever adding, modifying, or reviewing any Copilot, agent, LLM, AI, tool, permission, sandbox, MCP, model, telemetry, feature-gate, setting, configuration, or enterprise control—especially anything an organization or administrator may need to manage. Start here to decide whether it belongs in runtime managed settings, a typed SDK contract, VS Code configuration policy, extension policy, or a split implementation. Run on every new Copilot/agent/LLM control and ANY change that adds a `policy:` field.

Quality

87%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

78%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is an exemplary lean routing index: a complete decision table, concrete commands and identifiers, and explicit validation requirements with no wasted tokens. Its one material flaw is that every one of its seven referenced guide files is absent from the bundle, which undermines the progressive-disclosure structure it is built around.

Suggestions

Ship the seven referenced guides (sdk-runtime-policy.md, vscode-policy.md, extension-policy.md, legacy-permission-policy.md, github-managed-settings.md, policy-support-maintenance.md, local-testing.md) in the skill bundle — every link in the routing table and supporting-references list currently resolves to a missing file.

Inline a minimal fallback for the most common route (e.g. a short checklist for adding a runtime managed setting) so the body remains actionable when the guides are unavailable.

Add a brief error-recovery note for the validation requirements (what to do when export-policy-data output or policy-refresh tests fail) to close the workflow feedback loop.

Dimension	Reasoning	Score
Conciseness	The ~30-line body is a dense routing table plus rules with no explanation of concepts Claude already knows and no padding; every line carries a routing decision, a hard rule, or a test requirement. The version-sensitive 'pre-1.133' references are confined to a bounded legacy-migration table row rather than inline date-conditional guidance, fitting the top anchor.	5 / 5
Actionability	Concrete and executable in the main: 'pnpm run export-policy-data', 'Never hand-edit build/lib/policies/policyData.jsonc', 'extensionConfigurationPolicy in vscode-distro', and a table that fully determines each routing decision. Not 5 because implementation detail for each destination is deferred to the linked guides, leaving minor gaps if read in isolation.	4 / 5
Workflow Clarity	The routing table gives an unambiguous decision sequence, and explicit checkpoints are present: 'needs an agentHostPolicySupport entry and pnpm run export-policy-data' and 'Test policy refresh/removal, keyboard and pointer activation, and stale menus'. Not 5 because there is no error-recovery feedback loop for when validation or tests fail.	4 / 5
Progressive Disclosure	References are clearly signaled, one level deep, and annotated (e.g. '[sdk-runtime-policy.md](./sdk-runtime-policy.md)', '[local-testing.md](./local-testing.md): mock policy server'), but none of the seven linked guide files exist in the skill bundle — no references/, scripts/, or assets/ directories and no root-level guides. All implementation detail is delegated to missing files, so the otherwise-excellent navigation structure is broken in practice.	3 / 5
	Total	16 / 20 Passed

Description

91%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong description: it states a concrete routing action, enumerates the exact destinations, and gives explicit and varied trigger conditions including a precise `policy:` field trigger. The only weaknesses are a very broad catch-all trigger surface and the absence of any detail about what happens after the routing decision.

Dimension	Reasoning	Score
Specificity	Quotes 'adding, modifying, or reviewing' and 'decide whether it belongs in runtime managed settings, a typed SDK contract, VS Code configuration policy, extension policy, or a split implementation' — several concrete actions with five named destinations. Not 5 because only the routing/decision action is detailed; what to do after routing is unspecified.	4 / 5
Completeness	Explicitly answers both: what ('Start here to decide whether it belongs in runtime managed settings, a typed SDK contract, VS Code configuration policy, extension policy, or a split implementation') and when ('Use whenever adding, modifying, or reviewing any... control' and 'Run on every new Copilot/agent/LLM control'). Matches the top anchor with concrete trigger phrases.	5 / 5
Trigger Term Quality	Quotes 'Copilot, agent, LLM, AI, tool, permission, sandbox, MCP, model, telemetry, feature-gate, setting, configuration, or enterprise control' plus the exact code-level trigger 'ANY change that adds a `policy:` field' — comprehensive natural-term coverage including synonyms and a precise field-level trigger users and reviewers would naturally match.	5 / 5
Distinctiveness Conflict Risk	'especially anything an organization or administrator may need to manage' carves a clear enterprise-policy niche with distinct triggers, but the sweep 'any... setting, configuration, or enterprise control' is broad enough to overlap with generic settings/configuration skills. Mostly distinct with minor overlap risk, so 4 rather than 5.	4 / 5
	Total	18 / 20 Passed

Validation

93%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
relative_links	Relative link issues: 7 missing, 1 suspicious	Warning

	Total	15 / 16 Passed

Repository: microsoft/vscode
Path: .github/skills/policy-and-managed-settings/SKILL.md
Revision: f357833

Reviewed: 6 days ago

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.

policy-and-managed-settings

Table of Contents

Is this your skill?