docs: use npx ai-devkit@latest in command examples

codeaholicguy · codeaholicguy · commit 67904a26b5cc · 2026-02-21T08:29:47.000+01:00
diff --git a/docs/ai/design/feature-init-template.md b/docs/ai/design/feature-init-template.md
@@ -66,7 +66,7 @@ description: Define the technical architecture, components, and data models
   - `installSkills(skills: TemplateSkill[]): Promise<SkillInstallResult[]>`
 - Request/response formats
   - CLI command
-    - `npx ai-devkit init --template <path>`
+    - `npx ai-devkit@latest init --template <path>`
     - Path resolution: absolute path as-is; relative path resolved from current working directory.
   - Sample YAML template
     ```yaml
diff --git a/docs/ai/design/feature-memory-search-table-output.md b/docs/ai/design/feature-memory-search-table-output.md
@@ -49,9 +49,9 @@ description: Define the technical architecture, components, and data models
   - Keep `memorySearchCommand(options)` unchanged.
 - Request/response formats
   - Existing command:
-    - `npx ai-devkit memory search --query "<query>" [--tags "..."] [--scope "..."] [--limit N]`
+    - `npx ai-devkit@latest memory search --query "<query>" [--tags "..."] [--scope "..."] [--limit N]`
   - New table mode:
-    - `npx ai-devkit memory search --query "<query>" --table`
+    - `npx ai-devkit@latest memory search --query "<query>" --table`
   - Table output columns:
     - `id`
     - `title`
diff --git a/docs/ai/design/feature-memory-update.md b/docs/ai/design/feature-memory-update.md
@@ -71,7 +71,7 @@ interface MemoryUpdateOptions {
 ### CLI Command
 
 ```
-npx ai-devkit memory update --id <uuid> [--title <title>] [--content <content>] [--tags <tags>] [--scope <scope>]
+npx ai-devkit@latest memory update --id <uuid> [--title <title>] [--content <content>] [--tags <tags>] [--scope <scope>]
 ```
 
 | Option | Required | Description |
diff --git a/docs/ai/design/feature-setup-wizard.md b/docs/ai/design/feature-setup-wizard.md
@@ -71,10 +71,10 @@ description: Define the technical architecture, components, and data models
   - `renderReport(report): void`
 - Request/response formats
   - CLI entry points
-    - `npx ai-devkit setup` (interactive wizard default)
-    - `npx ai-devkit setup --non-interactive --profile <name> --tools codex,claude --assets commands,skills`
-    - `npx ai-devkit setup --dry-run`
-    - `npx ai-devkit setup --doctor` (recommended extension for diagnostics)
+    - `npx ai-devkit@latest setup` (interactive wizard default)
+    - `npx ai-devkit@latest setup --non-interactive --profile <name> --tools codex,claude --assets commands,skills`
+    - `npx ai-devkit@latest setup --dry-run`
+    - `npx ai-devkit@latest setup --doctor` (recommended extension for diagnostics)
   - Output
     - Human-readable summary by default.
     - Optional machine-readable JSON report (`--json`) for CI/auditing.
diff --git a/docs/ai/design/feature-skill-find.md b/docs/ai/design/feature-skill-find.md
@@ -58,7 +58,7 @@ description: Define the technical architecture, components, and data models
   - `ensureIndex(registries): Index` (checks TTL, registry list hash, optional head hash)
   - `searchIndex(index, keyword): SkillEntry[]`
 - Request/response formats
-  - CLI: `npx ai-devkit skill find <keyword>`
+  - CLI: `npx ai-devkit@latest skill find <keyword>`
   - Output: table or list with `skillName` and `description`
 - Authentication/authorization approach
   - GitHub unauthenticated rate limits apply (60 requests/hour).
diff --git a/docs/ai/implementation/feature-lint-command.md b/docs/ai/implementation/feature-lint-command.md
@@ -66,7 +66,7 @@ description: Technical implementation notes, patterns, and code guidelines
 
 - Git command failures are converted into deterministic lint results (miss or warn), not thrown errors.
 - Missing files are reported with explicit path and remediation guidance.
-- Output includes suggested fixes (for example `npx ai-devkit init`, `git worktree add ...`).
+- Output includes suggested fixes (for example `npx ai-devkit@latest init`, `git worktree add ...`).
 
 ## Performance Considerations
 **How do we keep it fast?**
diff --git a/docs/ai/implementation/feature-web-skill-search.md b/docs/ai/implementation/feature-web-skill-search.md
@@ -52,7 +52,7 @@ const filteredSkills = skills.filter(skill =>
 
 ### Copy to Clipboard
 ```typescript
-navigator.clipboard.writeText(`npx ai-devkit skill add ${skill.name}`);
+navigator.clipboard.writeText(`npx ai-devkit@latest skill add ${skill.name}`);
 ```
 
 ## Integration Points
diff --git a/docs/ai/planning/feature-lint-command.md b/docs/ai/planning/feature-lint-command.md
@@ -31,7 +31,7 @@ description: Break down work into actionable tasks and estimate timeline
 ### Phase 3: Integration & Polish
 - [x] Task 3.1: Update help text and README command documentation
 - [x] Task 3.2: Decide whether to keep `skills/dev-lifecycle/scripts/check-docs.sh` as wrapper or migrate references to `ai-devkit lint`
-- [x] Task 3.3: Add actionable remediation guidance in failures (`npx ai-devkit init`, worktree creation command)
+- [x] Task 3.3: Add actionable remediation guidance in failures (`npx ai-devkit@latest init`, worktree creation command)
 - [x] Task 3.4: Validate behavior against existing lifecycle docs and feature naming conventions
 
 ### Phase 4: Validation
diff --git a/docs/ai/requirements/feature-init-template.md b/docs/ai/requirements/feature-init-template.md
@@ -10,7 +10,7 @@ description: Clarify the problem space, gather requirements, and define success
 
 **What problem are we solving?**
 
-- `npx ai-devkit init` currently requires interactive step-by-step prompts, which slows down repeatable setup and automation.
+- `npx ai-devkit@latest init` currently requires interactive step-by-step prompts, which slows down repeatable setup and automation.
 - Teams and power users cannot define a reusable bootstrap configuration that includes environments, phases, and skills in one command.
 - Skill installation during init is not fully declarative from a single template file.
 - Who is affected:
@@ -23,12 +23,12 @@ description: Clarify the problem space, gather requirements, and define success
 **What do we want to achieve?**
 
 - Primary goals
-  - Add support for `npx ai-devkit init --template <file>` (example: `npx ai-devkit init --template test.yaml`).
+  - Add support for `npx ai-devkit@latest init --template <file>` (example: `npx ai-devkit@latest init --template test.yaml`).
   - Allow template-driven initialization for:
     - `environments`
     - `skills` (`registry`, `skill`)
     - `phases`
-  - Automatically install requested skills via `npx ai-devkit skill add` during init.
+  - Automatically install requested skills via `npx ai-devkit@latest skill add` during init.
   - Reduce or eliminate interactive prompts when a valid template provides required values.
 - Secondary goals
   - Keep backward compatibility for current interactive `init` flow.
@@ -43,7 +43,7 @@ description: Clarify the problem space, gather requirements, and define success
 
 **How will users interact with the solution?**
 
-- As a developer, I want to run `npx ai-devkit init --template test.yaml` so that I can initialize without answering prompts one by one.
+- As a developer, I want to run `npx ai-devkit@latest init --template test.yaml` so that I can initialize without answering prompts one by one.
 - As a team lead, I want to commit a template file so that every teammate gets the same environments, skills, and phases.
 - As a CI maintainer, I want deterministic initialization from template so that pipeline setup is reproducible.
 - Key workflows and scenarios
@@ -70,7 +70,7 @@ description: Clarify the problem space, gather requirements, and define success
   - Template format supports both YAML (`.yml`/`.yaml`) and JSON (`.json`) in v1.
   - Template path accepts both relative and absolute file paths.
   - Init parses template and applies `environments`, `skills`, and `phases`.
-  - For each template skill entry, init triggers the same install path as `npx ai-devkit skill add`.
+  - For each template skill entry, init triggers the same install path as `npx ai-devkit@latest skill add`.
   - Multiple entries with the same registry and different skill names are all processed.
   - Skill installation continues after individual failures and reports all failed items in the final summary.
   - If one or more skills fail, command exits with status code `0` and warning output.
diff --git a/docs/ai/requirements/feature-lint-command.md b/docs/ai/requirements/feature-lint-command.md
@@ -72,7 +72,7 @@ description: Clarify the problem space, gather requirements, and define success
   - `ai-devkit lint --feature <name>` checks dedicated worktree/branch convention (`feature-<name>`).
   - Missing dedicated worktree returns a warning, not a hard failure.
   - Command exits `0` when all required checks pass and non-zero when required checks fail.
-  - Output includes recommended remediation (for example `npx ai-devkit init` or worktree creation commands).
+  - Output includes recommended remediation (for example `npx ai-devkit@latest init` or worktree creation commands).
   - `--json` returns machine-readable structured results for CI tooling.
 - Performance benchmarks (if applicable)
   - File-system checks and git checks should avoid expensive scans and run in sub-second to low-second range.
diff --git a/docs/ai/requirements/feature-memory-search-table-output.md b/docs/ai/requirements/feature-memory-search-table-output.md
@@ -38,9 +38,9 @@ description: Clarify the problem space, gather requirements, and define success
 ## Success Criteria
 **How will we know when we're done?**
 
-- `npx ai-devkit memory search --query "<q>" --table` displays a terminal table.
+- `npx ai-devkit@latest memory search --query "<q>" --table` displays a terminal table.
 - Table columns are present and ordered: `id`, `title`, `scope`.
-- `npx ai-devkit memory search --query "<q>"` without `--table` preserves current JSON output behavior.
+- `npx ai-devkit@latest memory search --query "<q>"` without `--table` preserves current JSON output behavior.
 - Existing search filters (`--tags`, `--scope`, `--limit`) continue to work with `--table`.
 - When no results are found in table mode, CLI prints a clear warning and exits successfully (exit code `0`).
 
diff --git a/docs/ai/requirements/feature-memory-update.md b/docs/ai/requirements/feature-memory-update.md
@@ -10,14 +10,14 @@ description: Allow users to update existing memory content by ID via the CLI
 
 Users can store and search knowledge items via the CLI (`memory store`, `memory search`), but there is no way to update an existing entry. If a knowledge item becomes outdated or contains errors, the user cannot modify it — they would need to delete and re-create it (and deletion isn't supported either). This creates stale knowledge that can mislead AI assistants relying on the memory service.
 
-- **Who is affected?** Developers and AI assistants using `npx ai-devkit memory` to manage knowledge.
+- **Who is affected?** Developers and AI assistants using `npx ai-devkit@latest memory` to manage knowledge.
 - **Current workaround:** None — users cannot modify stored knowledge at all.
 
 ## Goals & Objectives
 
 **Primary goals:**
 - Allow users to update the title, content, tags, and/or scope of an existing memory item by its ID
-- Expose the update operation via CLI (`npx ai-devkit memory update`)
+- Expose the update operation via CLI (`npx ai-devkit@latest memory update`)
 - Expose the update operation via MCP tool (`memory.updateKnowledge`)
 
 **Non-goals:**
@@ -44,7 +44,7 @@ Users can store and search knowledge items via the CLI (`memory store`, `memory
 
 ## Success Criteria
 
-1. `npx ai-devkit memory update --id <uuid> [options]` successfully updates the specified fields
+1. `npx ai-devkit@latest memory update --id <uuid> [options]` successfully updates the specified fields
 2. Only provided fields are updated; unspecified fields remain unchanged
 3. Validation rules (title length, content length, tag format, scope format) are enforced on updated values
 4. Duplicate detection (normalized title + scope, content hash + scope) is enforced, excluding the item being updated
diff --git a/docs/ai/requirements/feature-setup-wizard.md b/docs/ai/requirements/feature-setup-wizard.md
@@ -25,7 +25,7 @@ description: Clarify the problem space, gather requirements, and define success
 **What do we want to achieve?**
 
 - Primary goals
-  - Turn `npx ai-devkit setup` into a guided interactive wizard by default (no required `--global` gate).
+  - Turn `npx ai-devkit@latest setup` into a guided interactive wizard by default (no required `--global` gate).
   - Configure local global environments for supported tools in one run (for example: Codex, Claude Code, Antigravity).
   - Support setup of global assets: commands, skills, and instruction/context files where each tool supports them.
   - Make setup idempotent, re-runnable, and safe (preview + backup + overwrite policy).
@@ -49,8 +49,8 @@ description: Clarify the problem space, gather requirements, and define success
 - As a power user, I want `--dry-run`, diff preview, and non-interactive flags so that I can automate setup safely.
 - As a returning user, I want re-running setup to be fast and safe so that I can update assets without breaking my custom files.
 - Key workflows and scenarios
-  - Run `npx ai-devkit setup` -> choose persona/profile -> select tools -> choose assets -> preview -> apply -> report.
-  - Run `npx ai-devkit setup --non-interactive --profile team-default` for deterministic onboarding.
+  - Run `npx ai-devkit@latest setup` -> choose persona/profile -> select tools -> choose assets -> preview -> apply -> report.
+  - Run `npx ai-devkit@latest setup --non-interactive --profile team-default` for deterministic onboarding.
   - Run setup again after upgrading ai-devkit -> detect drift -> apply selective updates.
 - Edge cases to consider
   - Tool installed but config directory missing.
@@ -68,7 +68,7 @@ description: Clarify the problem space, gather requirements, and define success
   - 90%+ of sampled users can complete setup without external docs, measured via onboarding study or opt-in post-setup prompt feedback.
   - Re-run setup changes only drifted files (idempotent behavior).
 - Acceptance criteria
-  - `npx ai-devkit setup` launches wizard by default.
+  - `npx ai-devkit@latest setup` launches wizard by default.
   - Wizard supports at least Codex, Claude Code, and Antigravity in v1.
   - User can select assets by type (`commands`, `skills`, instruction docs) per tool.
   - Wizard shows planned file operations before writing.
diff --git a/docs/ai/requirements/feature-skill-find.md b/docs/ai/requirements/feature-skill-find.md
@@ -16,7 +16,7 @@ description: Clarify the problem space, gather requirements, and define success
 ## Goals & Objectives
 **What do we want to achieve?**
 
-- Provide `npx ai-devkit skill find <keyword>` to list matching skills.
+- Provide `npx ai-devkit@latest skill find <keyword>` to list matching skills.
 - Make search fast without cloning full registries.
 - Keep results reasonably fresh via a lightweight index update trigger.
 - Keep output useful for install (skill name + description).
@@ -34,7 +34,7 @@ description: Clarify the problem space, gather requirements, and define success
 ## Success Criteria
 **How will we know when we're done?**
 
-- `npx ai-devkit skill find typescript` returns matching skills within 500ms with warm cache.
+- `npx ai-devkit@latest skill find typescript` returns matching skills within 500ms with warm cache.
 - Index update does not clone full registries for typical usage.
 - Search works across all registries in `skills/registry.json`.
 - Clear output format and exit codes for no matches or errors.
diff --git a/docs/ai/testing/feature-memory-search-table-output.md b/docs/ai/testing/feature-memory-search-table-output.md
@@ -40,8 +40,8 @@ description: Define testing approach, test cases, and quality assurance
 ## End-to-End Tests
 **What user flows need validation?**
 
-- [ ] Manual smoke: run `npx ai-devkit memory search --query "<q>"`.
-- [ ] Manual smoke: run `npx ai-devkit memory search --query "<q>" --table`.
+- [ ] Manual smoke: run `npx ai-devkit@latest memory search --query "<q>"`.
+- [ ] Manual smoke: run `npx ai-devkit@latest memory search --query "<q>" --table`.
 
 ## Test Data
 **What data do we use for testing?**
diff --git a/docs/ai/testing/feature-setup-wizard.md b/docs/ai/testing/feature-setup-wizard.md
@@ -71,9 +71,9 @@ description: Define testing approach, test cases, and quality assurance
 - Browser/device compatibility
   - Not applicable
 - Smoke tests after deployment
-  - `npx ai-devkit setup` (interactive)
-  - `npx ai-devkit setup --dry-run`
-  - `npx ai-devkit setup --non-interactive --profile quickstart --tools codex,claude`
+  - `npx ai-devkit@latest setup` (interactive)
+  - `npx ai-devkit@latest setup --dry-run`
+  - `npx ai-devkit@latest setup --non-interactive --profile quickstart --tools codex,claude`
 
 ## Performance Testing
 **How do we validate performance?**
diff --git a/docs/ai/testing/feature-skill-find.md b/docs/ai/testing/feature-skill-find.md
@@ -70,7 +70,7 @@ description: Define testing approach, test cases, and quality assurance
 - Browser/device compatibility
   - Not applicable
 - Smoke tests after deployment
-  - Run `npx ai-devkit skill find typescript`
+  - Run `npx ai-devkit@latest skill find typescript`
 
 ## Performance Testing
 **How do we validate performance?**
