@alucify/cli 0.6.5 → 0.7.1
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/dist/cli.js +1590 -1061
- package/dist/docs/help/cli/_meta.json +1 -0
- package/dist/docs/help/cli/authentication.md +70 -0
- package/dist/docs/help/cli/automation.md +72 -0
- package/dist/docs/help/cli/commands/_meta.json +1 -0
- package/dist/docs/help/cli/commands/add-spec.md +49 -0
- package/dist/docs/help/cli/commands/analyze-code.md +54 -0
- package/dist/docs/help/cli/commands/analyze-progress.md +73 -0
- package/dist/docs/help/cli/commands/analyze-specs.md +60 -0
- package/dist/docs/help/cli/commands/capture-intent.md +56 -0
- package/dist/docs/help/cli/commands/check.md +49 -0
- package/dist/docs/help/cli/commands/context.md +52 -0
- package/dist/docs/help/cli/commands/continue.md +40 -0
- package/dist/docs/help/cli/commands/enable-disable.md +53 -0
- package/dist/docs/help/cli/commands/help.md +30 -0
- package/dist/docs/help/cli/commands/mcp-install.md +54 -0
- package/dist/docs/help/cli/commands/plan.md +43 -0
- package/dist/docs/help/cli/commands/progress.md +58 -0
- package/dist/docs/help/cli/commands/status.md +50 -0
- package/dist/docs/help/cli/commands/ui.md +50 -0
- package/dist/docs/help/cli/commands/usage.md +65 -0
- package/dist/docs/help/cli/commands/version.md +66 -0
- package/dist/docs/help/cli/installation.md +70 -0
- package/dist/docs/help/cli/quickstart.md +75 -0
- package/dist/docs/help/cli/ui/_meta.json +1 -0
- package/dist/docs/help/cli/ui/ai-coding-risks-tab.md +46 -0
- package/dist/docs/help/cli/ui/appgraph-visualizer-tab.md +54 -0
- package/dist/docs/help/cli/ui/artifacts-sidebar.md +53 -0
- package/dist/docs/help/cli/ui/chat-panel.md +56 -0
- package/dist/docs/help/cli/ui/code-analysis-tab.md +36 -0
- package/dist/docs/help/cli/ui/dashboard.md +44 -0
- package/dist/docs/help/cli/ui/launching.md +46 -0
- package/dist/docs/help/cli/ui/requirements-progress-tab.md +49 -0
- package/dist/docs/help/cli/ui/settings-tab.md +52 -0
- package/dist/docs/help/cli/ui/tech-debt-tab.md +29 -0
- package/dist/docs/help/cli/ui/token-usage-tab.md +46 -0
- package/dist/docs/help/concepts/_meta.json +1 -0
- package/dist/docs/help/concepts/ai-coding-risks.md +50 -0
- package/dist/docs/help/concepts/ai-readiness.md +34 -0
- package/dist/docs/help/concepts/analysis-modes.md +66 -0
- package/dist/docs/help/concepts/appgraph.md +56 -0
- package/dist/docs/help/concepts/artifacts-and-specs.md +46 -0
- package/dist/docs/help/concepts/authentication-modes.md +68 -0
- package/dist/docs/help/concepts/blast-radius.md +37 -0
- package/dist/docs/help/concepts/chat-refinement.md +40 -0
- package/dist/docs/help/concepts/congruency-coverage-completeness.md +50 -0
- package/dist/docs/help/concepts/implementation-progress.md +44 -0
- package/dist/docs/help/concepts/intent-capture.md +46 -0
- package/dist/docs/help/concepts/tech-debt-analysis.md +65 -0
- package/dist/docs/help/concepts/token-usage.md +70 -0
- package/dist/docs/help/concepts/versions-and-snapshots.md +57 -0
- package/dist/docs/help/concepts/workflow-overview.md +40 -0
- package/dist/docs/help/desktop/_meta.json +1 -0
- package/dist/docs/help/desktop/installation.md +56 -0
- package/dist/docs/help/desktop/keyboard-shortcuts.md +47 -0
- package/dist/docs/help/desktop/menu-bar.md +47 -0
- package/dist/docs/help/desktop/notifications.md +44 -0
- package/dist/docs/help/desktop/quickstart.md +43 -0
- package/dist/docs/help/desktop/switching-workspaces.md +39 -0
- package/dist/docs/help/desktop/workspace-picker.md +38 -0
- package/dist/docs/help/extension/_meta.json +1 -0
- package/dist/docs/help/extension/authentication/_meta.json +1 -0
- package/dist/docs/help/extension/authentication/api-key-setup.md +47 -0
- package/dist/docs/help/extension/authentication/claude-code-setup.md +43 -0
- package/dist/docs/help/extension/authentication/signing-in.md +36 -0
- package/dist/docs/help/extension/authentication/switching-auth-modes.md +28 -0
- package/dist/docs/help/extension/chat/_meta.json +1 -0
- package/dist/docs/help/extension/chat/asking-about-your-project.md +39 -0
- package/dist/docs/help/extension/code-analysis/_meta.json +1 -0
- package/dist/docs/help/extension/code-analysis/analysis-modes.md +63 -0
- package/dist/docs/help/extension/code-analysis/running-code-analysis.md +54 -0
- package/dist/docs/help/extension/code-analysis/understanding-ai-coding-risks.md +75 -0
- package/dist/docs/help/extension/code-analysis/understanding-tech-debt-analysis.md +90 -0
- package/dist/docs/help/extension/code-analysis/understanding-the-code-analysis-tab.md +49 -0
- package/dist/docs/help/extension/dev-progress/_meta.json +1 -0
- package/dist/docs/help/extension/dev-progress/analyzing-commits.md +56 -0
- package/dist/docs/help/extension/dev-progress/understanding-the-dev-progress-tab.md +40 -0
- package/dist/docs/help/extension/installation.md +34 -0
- package/dist/docs/help/extension/intent-capture/_meta.json +1 -0
- package/dist/docs/help/extension/intent-capture/enabling-intent-capture.md +40 -0
- package/dist/docs/help/extension/intent-capture/understanding-intent-capture.md +40 -0
- package/dist/docs/help/extension/quickstart.md +60 -0
- package/dist/docs/help/extension/resetting-analysis.md +46 -0
- package/dist/docs/help/extension/settings/_meta.json +1 -0
- package/dist/docs/help/extension/settings/notification-preferences.md +38 -0
- package/dist/docs/help/extension/settings/telemetry.md +45 -0
- package/dist/docs/help/extension/spec-analysis/_meta.json +1 -0
- package/dist/docs/help/extension/spec-analysis/adding-specs.md +40 -0
- package/dist/docs/help/extension/spec-analysis/analyzing-specs.md +60 -0
- package/dist/docs/help/extension/spec-analysis/auditing-specs.md +30 -0
- package/dist/docs/help/extension/spec-analysis/filtering-and-browsing-specs.md +43 -0
- package/dist/docs/help/extension/spec-analysis/generating-specs.md +47 -0
- package/dist/docs/help/extension/spec-analysis/refining-specs.md +53 -0
- package/dist/docs/help/extension/spec-analysis/understanding-the-spec-analysis-tab.md +55 -0
- package/dist/docs/help/extension/token-usage/_meta.json +1 -0
- package/dist/docs/help/extension/token-usage/understanding-token-usage.md +64 -0
- package/dist/docs/help/extension/token-usage/viewing-token-usage.md +63 -0
- package/dist/docs/help/faq.md +128 -0
- package/dist/docs/help/getting-started/_meta.json +1 -0
- package/dist/docs/help/getting-started/choosing-a-ui.md +60 -0
- package/dist/docs/help/getting-started/first-analysis.md +86 -0
- package/dist/docs/help/getting-started/installation.md +65 -0
- package/dist/docs/help/getting-started/key-concepts.md +73 -0
- package/dist/docs/help/getting-started/supported-project-types.md +65 -0
- package/dist/docs/help/getting-started/system-requirements.md +61 -0
- package/dist/docs/help/integrations/_meta.json +1 -0
- package/dist/docs/help/integrations/ci-cd.md +95 -0
- package/dist/docs/help/integrations/claude-code-hooks.md +54 -0
- package/dist/docs/help/integrations/claude-code.md +123 -0
- package/dist/docs/help/integrations/git-hooks.md +64 -0
- package/dist/docs/help/mcp-server/_meta.json +1 -0
- package/dist/docs/help/mcp-server/quickstart.md +59 -0
- package/dist/docs/help/mcp-server/resources/_meta.json +1 -0
- package/dist/docs/help/mcp-server/resources/terminology-resource.md +74 -0
- package/dist/docs/help/mcp-server/setup.md +75 -0
- package/dist/docs/help/mcp-server/tools/_meta.json +1 -0
- package/dist/docs/help/mcp-server/tools/coverage-gaps.md +58 -0
- package/dist/docs/help/mcp-server/tools/find-similar-context.md +65 -0
- package/dist/docs/help/mcp-server/tools/get-analysis.md +54 -0
- package/dist/docs/help/mcp-server/tools/get-details.md +48 -0
- package/dist/docs/help/mcp-server/tools/impact-analysis.md +49 -0
- package/dist/docs/help/mcp-server/tools/implementation-tasks.md +68 -0
- package/dist/docs/help/mcp-server/tools/link-node.md +51 -0
- package/dist/docs/help/mcp-server/tools/query.md +53 -0
- package/dist/docs/help/mcp-server/tools/tools-reference.md +63 -0
- package/dist/docs/help/mcp-server/tools/verify-plan.md +82 -0
- package/dist/docs/help/mcp-server/tools/version-diff.md +42 -0
- package/dist/docs/help/mcp-server/using-with-claude-code.md +72 -0
- package/dist/docs/help/reference/_meta.json +1 -0
- package/dist/docs/help/reference/alucify-directory.md +86 -0
- package/dist/docs/help/reference/file-types-and-formats.md +76 -0
- package/dist/docs/help/reference/glossary.md +60 -0
- package/dist/docs/help/reference/how-claude-code-integration-works.md +123 -0
- package/dist/docs/help/reference/ui-parity-matrix.md +121 -0
- package/dist/docs/help/vscode/_meta.json +1 -0
- package/dist/docs/help/vscode/advanced/_meta.json +1 -0
- package/dist/docs/help/vscode/advanced/congruency-analysis.md +32 -0
- package/dist/docs/help/vscode/advanced/coverage-analysis.md +31 -0
- package/dist/docs/help/vscode/advanced/developer-mode.md +34 -0
- package/dist/docs/help/vscode/advanced/findings-synthesis.md +27 -0
- package/dist/docs/help/vscode/advanced/impact-analysis.md +31 -0
- package/dist/docs/help/vscode/advanced/reset-extension.md +39 -0
- package/dist/docs/help/vscode/appgraph-visualizer/_meta.json +1 -0
- package/dist/docs/help/vscode/appgraph-visualizer/filtering-and-search.md +63 -0
- package/dist/docs/help/vscode/appgraph-visualizer/navigating-the-graph.md +34 -0
- package/dist/docs/help/vscode/appgraph-visualizer/node-details.md +49 -0
- package/dist/docs/help/vscode/appgraph-visualizer/opening-the-visualizer.md +33 -0
- package/dist/docs/help/vscode/authentication/_meta.json +1 -0
- package/dist/docs/help/vscode/authentication/api-key-setup.md +47 -0
- package/dist/docs/help/vscode/authentication/claude-code-setup.md +43 -0
- package/dist/docs/help/vscode/authentication/signing-in.md +36 -0
- package/dist/docs/help/vscode/authentication/switching-auth-modes.md +28 -0
- package/dist/docs/help/vscode/chat/_meta.json +1 -0
- package/dist/docs/help/vscode/chat/asking-about-your-project.md +39 -0
- package/dist/docs/help/vscode/chat/generating-artifacts-in-chat.md +53 -0
- package/dist/docs/help/vscode/chat/refining-specs-in-chat.md +47 -0
- package/dist/docs/help/vscode/code-analysis/_meta.json +1 -0
- package/dist/docs/help/vscode/code-analysis/ai-coding-risks-panel.md +42 -0
- package/dist/docs/help/vscode/code-analysis/project-analysis-panel.md +46 -0
- package/dist/docs/help/vscode/code-analysis/running-code-analysis.md +54 -0
- package/dist/docs/help/vscode/code-analysis/tech-debt-panel.md +27 -0
- package/dist/docs/help/vscode/command-palette.md +83 -0
- package/dist/docs/help/vscode/dev-progress/_meta.json +1 -0
- package/dist/docs/help/vscode/dev-progress/analyzing-commits.md +56 -0
- package/dist/docs/help/vscode/dev-progress/progress-panel-actions.md +56 -0
- package/dist/docs/help/vscode/dev-progress/understanding-the-dev-progress-tab.md +40 -0
- package/dist/docs/help/vscode/installation.md +34 -0
- package/dist/docs/help/vscode/intent-capture/_meta.json +1 -0
- package/dist/docs/help/vscode/intent-capture/analyzing-intent.md +49 -0
- package/dist/docs/help/vscode/intent-capture/enabling-intent-capture.md +40 -0
- package/dist/docs/help/vscode/quickstart.md +60 -0
- package/dist/docs/help/vscode/resetting-analysis.md +46 -0
- package/dist/docs/help/vscode/settings/_meta.json +1 -0
- package/dist/docs/help/vscode/settings/advanced-mode.md +53 -0
- package/dist/docs/help/vscode/settings/all-settings.md +66 -0
- package/dist/docs/help/vscode/settings/auto-analyze-specs.md +43 -0
- package/dist/docs/help/vscode/settings/notification-preferences.md +38 -0
- package/dist/docs/help/vscode/settings/telemetry.md +45 -0
- package/dist/docs/help/vscode/spec-analysis/_meta.json +1 -0
- package/dist/docs/help/vscode/spec-analysis/adding-specs.md +40 -0
- package/dist/docs/help/vscode/spec-analysis/analyzing-specs.md +60 -0
- package/dist/docs/help/vscode/spec-analysis/auditing-specs.md +30 -0
- package/dist/docs/help/vscode/spec-analysis/filtering-and-browsing-specs.md +43 -0
- package/dist/docs/help/vscode/spec-analysis/generating-specs.md +47 -0
- package/dist/docs/help/vscode/spec-analysis/refining-specs.md +53 -0
- package/dist/docs/help/vscode/spec-analysis/specifications-panel.md +46 -0
- package/dist/docs/help/vscode/status-bar.md +44 -0
- package/dist/docs/help/vscode/token-usage/_meta.json +1 -0
- package/dist/docs/help/vscode/token-usage/viewing-token-usage.md +63 -0
- package/dist/docs/help/vscode/versions/_meta.json +1 -0
- package/dist/docs/help/vscode/versions/comparing-versions.md +47 -0
- package/dist/docs/help/vscode/versions/deleting-versions.md +44 -0
- package/dist/docs/help/vscode/versions/switching-versions.md +43 -0
- package/dist/docs/help/vscode/workflow-controls.md +49 -0
- package/dist/docs/help/what-is-alucify.md +56 -0
- package/dist/mcp-server-vendored.js +28 -28
- package/dist/web/app.css +38 -0
- package/dist/web/app.js +671 -0
- package/dist/web/favicon.png +0 -0
- package/dist/web/favicon.svg +7 -0
- package/dist/web/index.html +15 -0
- package/package.json +1 -1
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{ "title": "Advanced", "order": 10 }
|
|
@@ -0,0 +1,32 @@
|
|
|
1
|
+
---
|
|
2
|
+
order: 1
|
|
3
|
+
title: "Congruency Analysis"
|
|
4
|
+
description: "Running congruency analysis as a standalone pass."
|
|
5
|
+
---
|
|
6
|
+
|
|
7
|
+
# Congruency Analysis
|
|
8
|
+
|
|
9
|
+
Congruency analysis checks how well your specs and code agree. It's normally run automatically as part of [spec analysis](../spec-analysis/analyzing-specs.md), but Advanced Mode exposes it as a standalone command.
|
|
10
|
+
|
|
11
|
+
For the concept, see [Congruency, Coverage & Completeness](../../concepts/congruency-coverage-completeness.md).
|
|
12
|
+
|
|
13
|
+
## When to Run It Alone
|
|
14
|
+
|
|
15
|
+
- You've only updated specs (not code) and want a fast congruency refresh without re-running the whole spec analysis pipeline
|
|
16
|
+
- You want to try a different analysis mode just for congruency
|
|
17
|
+
- You're debugging unexpected congruency scores
|
|
18
|
+
|
|
19
|
+
## How to Run
|
|
20
|
+
|
|
21
|
+
- **Command Palette** — **Alucify (Advanced): Analyze Congruency** (requires [Advanced Mode](../settings/advanced-mode.md))
|
|
22
|
+
|
|
23
|
+
Pick an [analysis mode](../../concepts/analysis-modes.md) and the target specs when prompted.
|
|
24
|
+
|
|
25
|
+
## Output
|
|
26
|
+
|
|
27
|
+
A congruency-only report written to the current version. Scores are updated in the [Specifications Panel](../spec-analysis/specifications-panel.md) when done.
|
|
28
|
+
|
|
29
|
+
## Caveats
|
|
30
|
+
|
|
31
|
+
- Congruency depends on coverage data — if coverage hasn't been computed, results will be incomplete
|
|
32
|
+
- For the full view (congruency + coverage + completeness + risks + recommendations), run [Analyze Specs](../spec-analysis/analyzing-specs.md) instead
|
|
@@ -0,0 +1,31 @@
|
|
|
1
|
+
---
|
|
2
|
+
order: 2
|
|
3
|
+
title: "Coverage Analysis"
|
|
4
|
+
description: "Running coverage analysis as a standalone pass."
|
|
5
|
+
---
|
|
6
|
+
|
|
7
|
+
# Coverage Analysis
|
|
8
|
+
|
|
9
|
+
Coverage analysis identifies which parts of your codebase have no corresponding specification. Advanced Mode exposes it as a standalone command.
|
|
10
|
+
|
|
11
|
+
For the concept, see [Congruency, Coverage & Completeness](../../concepts/congruency-coverage-completeness.md).
|
|
12
|
+
|
|
13
|
+
## When to Run It Alone
|
|
14
|
+
|
|
15
|
+
- You've added new code and want a fast coverage refresh without full spec analysis
|
|
16
|
+
- You're triaging "missing spec" risks and want the latest coverage data
|
|
17
|
+
|
|
18
|
+
## How to Run
|
|
19
|
+
|
|
20
|
+
- **Command Palette** — **Alucify (Advanced): Analyze Coverage** (requires [Advanced Mode](../settings/advanced-mode.md))
|
|
21
|
+
|
|
22
|
+
Pick an [analysis mode](../../concepts/analysis-modes.md).
|
|
23
|
+
|
|
24
|
+
## Output
|
|
25
|
+
|
|
26
|
+
A coverage-only report. Coverage bars in the [Specifications Panel](../spec-analysis/specifications-panel.md) update when done.
|
|
27
|
+
|
|
28
|
+
## Related
|
|
29
|
+
|
|
30
|
+
- [AI Coding Risks](../../concepts/ai-coding-risks.md) — low coverage produces Missing UX / Data / Logic Spec risks
|
|
31
|
+
- [`get_remaining_work`](../../mcp-server/tools/coverage-gaps.md) — query coverage gaps from Claude Code
|
|
@@ -0,0 +1,34 @@
|
|
|
1
|
+
---
|
|
2
|
+
order: 6
|
|
3
|
+
title: "Developer Mode"
|
|
4
|
+
description: "Internal diagnostics for debugging the Alucify extension."
|
|
5
|
+
---
|
|
6
|
+
|
|
7
|
+
# Developer Mode
|
|
8
|
+
|
|
9
|
+
Developer Mode enables diagnostics used by the Alucify team (or contributors) when debugging the extension itself. You probably don't need this.
|
|
10
|
+
|
|
11
|
+
## What It Enables
|
|
12
|
+
|
|
13
|
+
- **Output Channel** — `Alucify` in VS Code's Output panel, with verbose logging
|
|
14
|
+
- **Extra panels** — internal state inspectors for workflow engine, agent runner, and MCP client
|
|
15
|
+
- **Override flags** — `alucify.agentsInternalPath` starts being respected (normally ignored)
|
|
16
|
+
|
|
17
|
+
## Turning It On
|
|
18
|
+
|
|
19
|
+
Developer Mode isn't a single setting — it's enabled when you set `alucify.agentsInternalPath` to a non-empty value (typically a path to the `alucify-agents-internal` repo for internal development).
|
|
20
|
+
|
|
21
|
+
For external users this is almost never what you want. If you're debugging behavior, start with:
|
|
22
|
+
- [Output Channel](../../reference/alucify-directory.md#debug-logs) — inspect logs without Developer Mode
|
|
23
|
+
- [Reset Extension](reset-extension.md) — clear state
|
|
24
|
+
- File an issue with reproduction steps
|
|
25
|
+
|
|
26
|
+
## Agent Path Override
|
|
27
|
+
|
|
28
|
+
`alucify.agentsInternalPath` lets internal developers point at a local copy of the agent definitions repo. Setting this to an empty string (default) uses the agents bundled with the extension.
|
|
29
|
+
|
|
30
|
+
Invalid paths fall back to bundled agents with a warning in the Output Channel.
|
|
31
|
+
|
|
32
|
+
## Reverting
|
|
33
|
+
|
|
34
|
+
Clear the setting (set to empty string). The extension goes back to normal user mode.
|
|
@@ -0,0 +1,27 @@
|
|
|
1
|
+
---
|
|
2
|
+
order: 4
|
|
3
|
+
title: "Findings Synthesis"
|
|
4
|
+
description: "Running quality insight synthesis as a standalone pass."
|
|
5
|
+
---
|
|
6
|
+
|
|
7
|
+
# Findings Synthesis
|
|
8
|
+
|
|
9
|
+
Findings Synthesis aggregates results from congruency, coverage, and impact analyses into a ranked list of actionable insights. It's the step that produces the Recommendations section in the Spec Analysis tab and the AI Coding Risks output.
|
|
10
|
+
|
|
11
|
+
## When to Run It Alone
|
|
12
|
+
|
|
13
|
+
- You've manually run congruency, coverage, or impact analyses separately and want to re-synthesize without a full spec analysis run
|
|
14
|
+
- You want to regenerate the Recommendations list after updating mitigations
|
|
15
|
+
|
|
16
|
+
## How to Run
|
|
17
|
+
|
|
18
|
+
- **Command Palette** — **Alucify (Advanced): Analyze Findings** (requires [Advanced Mode](../settings/advanced-mode.md))
|
|
19
|
+
|
|
20
|
+
## Output
|
|
21
|
+
|
|
22
|
+
- Updated recommendations in the Spec Analysis tab
|
|
23
|
+
- Refreshed AI Coding Risks (severities may shift based on new data)
|
|
24
|
+
|
|
25
|
+
## Prerequisites
|
|
26
|
+
|
|
27
|
+
This analysis reads results from prior congruency, coverage, and impact runs. If any of those are stale or missing, re-run them first.
|
|
@@ -0,0 +1,31 @@
|
|
|
1
|
+
---
|
|
2
|
+
order: 3
|
|
3
|
+
title: "Impact Analysis"
|
|
4
|
+
description: "Running blast-radius analysis as a standalone pass."
|
|
5
|
+
---
|
|
6
|
+
|
|
7
|
+
# Impact Analysis
|
|
8
|
+
|
|
9
|
+
Impact analysis computes blast radius — how much of your project is affected by changes in a given area.
|
|
10
|
+
|
|
11
|
+
For the concept, see [Blast Radius](../../concepts/blast-radius.md).
|
|
12
|
+
|
|
13
|
+
## When to Run It Alone
|
|
14
|
+
|
|
15
|
+
- You want to re-score blast radius after structural changes to the codebase
|
|
16
|
+
- You're investigating a specific area and want the latest propagation data
|
|
17
|
+
|
|
18
|
+
## How to Run
|
|
19
|
+
|
|
20
|
+
- **Command Palette** — **Alucify (Advanced): Analyze Impact** (requires [Advanced Mode](../settings/advanced-mode.md))
|
|
21
|
+
|
|
22
|
+
## Output
|
|
23
|
+
|
|
24
|
+
Updated blast-radius metrics per node. Shown in:
|
|
25
|
+
- Spec Analysis tab (project-level average)
|
|
26
|
+
- AppGraph Visualizer node details
|
|
27
|
+
- AI Coding Risks panel (High Blast Radius risks are re-scored)
|
|
28
|
+
|
|
29
|
+
## Related
|
|
30
|
+
|
|
31
|
+
- [`trace_dependencies`](../../mcp-server/tools/impact-analysis.md) — query impact from Claude Code
|
|
@@ -0,0 +1,39 @@
|
|
|
1
|
+
---
|
|
2
|
+
order: 5
|
|
3
|
+
title: "Reset Extension"
|
|
4
|
+
description: "Factory reset for the VS Code extension — clears auth and local state."
|
|
5
|
+
---
|
|
6
|
+
|
|
7
|
+
# Reset Extension
|
|
8
|
+
|
|
9
|
+
A full factory reset for the Alucify VS Code extension. Use this sparingly — it clears authentication and extension state, but does **not** delete your project's analysis data.
|
|
10
|
+
|
|
11
|
+
## What Reset Extension Does
|
|
12
|
+
|
|
13
|
+
- Clears the API key from VS Code secret storage
|
|
14
|
+
- Clears any Auth0 session tokens
|
|
15
|
+
- Resets extension-level state (welcome flags, auth-mode-chosen, panel states)
|
|
16
|
+
- Does **not** touch `.alucify/` in any project — analyses, specs, versions, and token logs are preserved
|
|
17
|
+
|
|
18
|
+
## How to Run
|
|
19
|
+
|
|
20
|
+
- **Command Palette** — **Alucify (Advanced): Reset Extension** (requires [Advanced Mode](../settings/advanced-mode.md))
|
|
21
|
+
|
|
22
|
+
A confirmation dialog appears warning what will be cleared. Click **Yes, reset** to proceed.
|
|
23
|
+
|
|
24
|
+
## After Reset
|
|
25
|
+
|
|
26
|
+
- The Welcome view reappears
|
|
27
|
+
- You'll be prompted to pick an auth mode again
|
|
28
|
+
- Analyses are untouched; reopen a project and its data is still there
|
|
29
|
+
|
|
30
|
+
## vs. Reset Analysis
|
|
31
|
+
|
|
32
|
+
- **[Reset Analysis](../resetting-analysis.md)** — archives a project's `.alucify/` data to start that project fresh; auth and extension settings are preserved
|
|
33
|
+
- **Reset Extension** — clears auth and extension settings; projects' analyses are preserved
|
|
34
|
+
|
|
35
|
+
## When to Use
|
|
36
|
+
|
|
37
|
+
- Debugging auth issues
|
|
38
|
+
- Switching to a different Anthropic account on this machine
|
|
39
|
+
- Reproducing a first-run experience
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{ "title": "AppGraph Visualizer", "order": 6 }
|
|
@@ -0,0 +1,63 @@
|
|
|
1
|
+
---
|
|
2
|
+
order: 3
|
|
3
|
+
title: "Filtering and Search"
|
|
4
|
+
description: "Narrowing the AppGraph view with filters and text search."
|
|
5
|
+
---
|
|
6
|
+
|
|
7
|
+
# Filtering and Search
|
|
8
|
+
|
|
9
|
+
The AppGraph Visualizer lets you narrow what's visible via filters (pinned in the left panel) and free-text search.
|
|
10
|
+
|
|
11
|
+
## Filter by Node Type
|
|
12
|
+
|
|
13
|
+
Toggle each node type on or off:
|
|
14
|
+
|
|
15
|
+
- **Interface** — UI screens, views, components
|
|
16
|
+
- **Logic** — functions, classes, modules
|
|
17
|
+
- **Schema** — data models, types, tables
|
|
18
|
+
- **FunctionalArea** — top-level feature areas
|
|
19
|
+
- **FeatureSpec** — feature definitions
|
|
20
|
+
- **AcceptanceCriterion** — testable conditions
|
|
21
|
+
- **Validation** — tests
|
|
22
|
+
- **Other** — anything else
|
|
23
|
+
|
|
24
|
+
See [AppGraph](../../concepts/appgraph.md) for what each type represents.
|
|
25
|
+
|
|
26
|
+
## Filter by Status
|
|
27
|
+
|
|
28
|
+
Narrow to specific [implementation statuses](../../concepts/implementation-progress.md):
|
|
29
|
+
|
|
30
|
+
- Planned
|
|
31
|
+
- In Progress
|
|
32
|
+
- Implemented
|
|
33
|
+
- Tested
|
|
34
|
+
- Deprecated
|
|
35
|
+
|
|
36
|
+
Useful for: "show me only what's still planned" or "what's in-progress right now?"
|
|
37
|
+
|
|
38
|
+
## Filter by Origin
|
|
39
|
+
|
|
40
|
+
Every node has an origin indicating where it came from:
|
|
41
|
+
|
|
42
|
+
- **Spec** — introduced by a specification artifact (with the spec name)
|
|
43
|
+
- **Code** — found during code analysis
|
|
44
|
+
- **Intent** — captured from a Claude Code session ([Intent Capture](../../concepts/intent-capture.md))
|
|
45
|
+
|
|
46
|
+
Filter to, for example, see only intent-captured nodes — useful to review what Claude Code sessions have introduced to your project.
|
|
47
|
+
|
|
48
|
+
## Filter by Neighborhood
|
|
49
|
+
|
|
50
|
+
Click a node and choose **Focus on Neighborhood** (N hops) to hide everything not reachable from this node within N edges. Good for: trace dependencies, understand a subsystem.
|
|
51
|
+
|
|
52
|
+
## Free-Text Search
|
|
53
|
+
|
|
54
|
+
The search bar at the top of the visualizer accepts any text. It matches node names, descriptions, and source file paths. Matches are highlighted; clicking a result focuses it.
|
|
55
|
+
|
|
56
|
+
## Combining Filters
|
|
57
|
+
|
|
58
|
+
Filters are AND-combined — checking "Interface" and "In Progress" shows only interface nodes that are in-progress. Text search intersects with that set.
|
|
59
|
+
|
|
60
|
+
## Clearing Filters
|
|
61
|
+
|
|
62
|
+
- **Reset all filters** — click the Reset button at the top of the filter panel
|
|
63
|
+
- **Clear search** — clear the search input
|
|
@@ -0,0 +1,34 @@
|
|
|
1
|
+
---
|
|
2
|
+
order: 2
|
|
3
|
+
title: "Navigating the Graph"
|
|
4
|
+
description: "Zooming, panning, and moving through the AppGraph Visualizer."
|
|
5
|
+
---
|
|
6
|
+
|
|
7
|
+
# Navigating the Graph
|
|
8
|
+
|
|
9
|
+
## Basic Navigation
|
|
10
|
+
|
|
11
|
+
- **Pan** — click and drag any empty space
|
|
12
|
+
- **Zoom** — scroll wheel (trackpad pinch on macOS)
|
|
13
|
+
- **Select a node** — single-click
|
|
14
|
+
- **Focus a node** — double-click to center and zoom in
|
|
15
|
+
- **Fit to screen** — click the Fit button in the toolbar
|
|
16
|
+
|
|
17
|
+
## Force-Directed Layout
|
|
18
|
+
|
|
19
|
+
The graph uses a force-directed layout — nodes repel each other and edges pull connected nodes together. The layout runs continuously by default, which is why nodes drift slowly until they settle.
|
|
20
|
+
|
|
21
|
+
- **Freeze layout** — click the lock icon in the toolbar to stop the simulation
|
|
22
|
+
- **Restart layout** — click again to resume
|
|
23
|
+
|
|
24
|
+
## Traversing Edges
|
|
25
|
+
|
|
26
|
+
Click any edge to traverse along it. The connected node becomes the new focus. Use this to explore dependency chains or implementation links.
|
|
27
|
+
|
|
28
|
+
## Mini-map
|
|
29
|
+
|
|
30
|
+
A mini-map in the bottom-right shows your current viewport relative to the full graph. Click any spot on the mini-map to jump there.
|
|
31
|
+
|
|
32
|
+
## Performance
|
|
33
|
+
|
|
34
|
+
Very large graphs (thousands of nodes) may render slowly with the force layout running. Filter the view first (see [Filtering and Search](filtering-and-search.md)) to narrow the scope, or freeze the layout once nodes have settled.
|
|
@@ -0,0 +1,49 @@
|
|
|
1
|
+
---
|
|
2
|
+
order: 4
|
|
3
|
+
title: "Node Details"
|
|
4
|
+
description: "Reading node information in the AppGraph Visualizer details pane."
|
|
5
|
+
---
|
|
6
|
+
|
|
7
|
+
# Node Details
|
|
8
|
+
|
|
9
|
+
When you click a node, a detail pane opens showing everything Alucify knows about it.
|
|
10
|
+
|
|
11
|
+
## What's Displayed
|
|
12
|
+
|
|
13
|
+
### Identity
|
|
14
|
+
- **Name** — canonical name of the node
|
|
15
|
+
- **Type** — node type (Logic, Interface, etc.)
|
|
16
|
+
- **ID** — internal AppGraph ID (copyable; used by MCP tools)
|
|
17
|
+
|
|
18
|
+
### Status
|
|
19
|
+
- **Status badge** — Planned / In Progress / Implemented / Tested / Deprecated
|
|
20
|
+
- **Gap notes** (for in-progress) — what's still missing
|
|
21
|
+
- **Status history** — status changes across versions
|
|
22
|
+
|
|
23
|
+
### Grounding
|
|
24
|
+
- **Source files** — where in the code this node lives (clickable; opens in editor)
|
|
25
|
+
- **Spec references** — which artifacts introduced or reference this node
|
|
26
|
+
- **Origin** — Spec / Code / Intent
|
|
27
|
+
|
|
28
|
+
### Relationships
|
|
29
|
+
- **Inbound edges** — what points at this node (callers, parents, etc.)
|
|
30
|
+
- **Outbound edges** — what this node points at (dependencies, children)
|
|
31
|
+
|
|
32
|
+
Click any edge entry to jump to the connected node.
|
|
33
|
+
|
|
34
|
+
### Metrics
|
|
35
|
+
- **[Blast radius](../../concepts/blast-radius.md)** — percentage of the project reachable from this node
|
|
36
|
+
- **Depth** — how many hops from the nearest root
|
|
37
|
+
|
|
38
|
+
## Jumping to Source
|
|
39
|
+
|
|
40
|
+
Clicking a source file opens it in the editor at the right location. For Logic nodes, this typically means the function definition; for Schema nodes, the type or table declaration.
|
|
41
|
+
|
|
42
|
+
## Editing
|
|
43
|
+
|
|
44
|
+
The visualizer is read-only. To change a spec, edit the spec file directly; to change code, edit the source. Alucify picks up the changes on the next analysis.
|
|
45
|
+
|
|
46
|
+
## Copying
|
|
47
|
+
|
|
48
|
+
- **Copy ID** — useful when asking Claude about a specific node via MCP tools
|
|
49
|
+
- **Copy as link** — creates a link you can paste into specs or docs that will navigate back to this node when clicked
|
|
@@ -0,0 +1,33 @@
|
|
|
1
|
+
---
|
|
2
|
+
order: 1
|
|
3
|
+
title: "Opening the Visualizer"
|
|
4
|
+
description: "How to open the AppGraph Visualizer panel in VS Code."
|
|
5
|
+
---
|
|
6
|
+
|
|
7
|
+
# Opening the AppGraph Visualizer
|
|
8
|
+
|
|
9
|
+
The AppGraph Visualizer is an interactive graph view of every node and edge in your [AppGraph](../../concepts/appgraph.md).
|
|
10
|
+
|
|
11
|
+
## Ways to Open It
|
|
12
|
+
|
|
13
|
+
- **Command Palette** — `Cmd+Shift+P` / `Ctrl+Shift+P` → search for **Alucify: Open AppGraph Visualizer**
|
|
14
|
+
- **Advanced menu** — in the Alucify Activity Bar view, expand **Advanced** and click **Open AppGraph Visualizer** (requires [Advanced Mode](../settings/advanced-mode.md) to be on)
|
|
15
|
+
- **Context links** — anywhere a node is displayed in the extension (e.g., the Specifications or AI Coding Risks panels), click the node to open the visualizer focused on it
|
|
16
|
+
|
|
17
|
+
## Prerequisites
|
|
18
|
+
|
|
19
|
+
You must have run at least one analysis — code or spec. Without an AppGraph, the visualizer is empty.
|
|
20
|
+
|
|
21
|
+
## Panel Behavior
|
|
22
|
+
|
|
23
|
+
The visualizer opens as a VS Code webview panel (not a sidebar view). It takes up the entire editor column by default but can be moved into a split view like any other VS Code tab.
|
|
24
|
+
|
|
25
|
+
## Reopening a Specific Node
|
|
26
|
+
|
|
27
|
+
The last node you clicked is remembered across sessions. When you reopen the visualizer, it focuses on that node automatically. To start fresh, clear all filters from the panel.
|
|
28
|
+
|
|
29
|
+
## See Also
|
|
30
|
+
|
|
31
|
+
- [Navigating the Graph](navigating-the-graph.md)
|
|
32
|
+
- [Filtering and Search](filtering-and-search.md)
|
|
33
|
+
- [Node Details](node-details.md)
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{ "title": "Authentication", "order": 3 }
|
|
@@ -0,0 +1,47 @@
|
|
|
1
|
+
---
|
|
2
|
+
order: 2
|
|
3
|
+
title: "Setting Up an Anthropic API Key"
|
|
4
|
+
description: "How to configure Alucify with your Anthropic API key for pay-per-use access."
|
|
5
|
+
---
|
|
6
|
+
|
|
7
|
+
# Setting Up an Anthropic API Key
|
|
8
|
+
|
|
9
|
+
Use your own Anthropic API key to pay per token for AI analysis.
|
|
10
|
+
|
|
11
|
+
## Getting an API Key
|
|
12
|
+
|
|
13
|
+
1. Go to [console.anthropic.com](https://console.anthropic.com)
|
|
14
|
+
2. Create an account or sign in
|
|
15
|
+
3. Navigate to **API Keys**
|
|
16
|
+
4. Click **Create Key** and copy the generated key (it starts with `sk-ant-`)
|
|
17
|
+
|
|
18
|
+
## Entering Your Key in Alucify
|
|
19
|
+
|
|
20
|
+
### During Welcome Setup
|
|
21
|
+
|
|
22
|
+
1. In Step 2 of the welcome screen, select **Anthropic API Key**
|
|
23
|
+
2. Enter your API key in the text field
|
|
24
|
+
3. Click **Save**
|
|
25
|
+
4. You'll see a confirmation: "Anthropic API key saved securely."
|
|
26
|
+
|
|
27
|
+
### After Setup
|
|
28
|
+
|
|
29
|
+
If you need to update your API key later:
|
|
30
|
+
|
|
31
|
+
1. Open the Command Palette (`Ctrl+Shift+P` / `Cmd+Shift+P`)
|
|
32
|
+
2. Search for **Alucify: Set Anthropic API Key**
|
|
33
|
+
3. Enter your new key
|
|
34
|
+
|
|
35
|
+
## Security
|
|
36
|
+
|
|
37
|
+
Your API key is stored securely in VS Code's built-in secret storage. It is never written to disk in plain text or included in any project files.
|
|
38
|
+
|
|
39
|
+
## Cost Considerations
|
|
40
|
+
|
|
41
|
+
API costs depend on your analysis mode and project size:
|
|
42
|
+
|
|
43
|
+
- **Speed mode** — Free (no AI calls, mechanical analysis only)
|
|
44
|
+
- **Balanced mode** — Moderate cost (uses Claude Sonnet)
|
|
45
|
+
- **Quality mode** — Higher cost (uses Claude Opus for comprehensive analysis)
|
|
46
|
+
|
|
47
|
+
The exact cost depends on your codebase size and the number of specs you analyze. For most projects, a single code analysis in Balanced mode costs a few dollars.
|
|
@@ -0,0 +1,43 @@
|
|
|
1
|
+
---
|
|
2
|
+
order: 3
|
|
3
|
+
title: "Using Claude Code / Claude Max Subscription"
|
|
4
|
+
description: "How to use your existing Claude Code or Claude Max subscription with Alucify."
|
|
5
|
+
---
|
|
6
|
+
|
|
7
|
+
# Using Claude Code / Claude Max Subscription
|
|
8
|
+
|
|
9
|
+
If you have a Claude Code or Claude Max subscription, you can use it with Alucify instead of paying per token with an API key.
|
|
10
|
+
|
|
11
|
+
## Prerequisites
|
|
12
|
+
|
|
13
|
+
1. **Claude Code CLI** installed on your machine — download from [claude.ai/download](https://claude.ai/download)
|
|
14
|
+
2. An active **Claude Code** or **Claude Max** subscription
|
|
15
|
+
3. The CLI authenticated (run `claude` in your terminal to verify)
|
|
16
|
+
|
|
17
|
+
## Setting Up
|
|
18
|
+
|
|
19
|
+
### During Welcome Setup
|
|
20
|
+
|
|
21
|
+
1. In Step 2 of the welcome screen, select **Claude Code Subscription**
|
|
22
|
+
2. If the Claude CLI is detected, you'll see a green status indicator
|
|
23
|
+
3. If the CLI is not detected, you'll see "(CLI not detected)" in red — install Claude Code first
|
|
24
|
+
|
|
25
|
+
### After Setup
|
|
26
|
+
|
|
27
|
+
You can switch to Claude Code mode at any time:
|
|
28
|
+
|
|
29
|
+
1. Open the Command Palette (`Ctrl+Shift+P` / `Cmd+Shift+P`)
|
|
30
|
+
2. Search for **Alucify: Switch Auth Mode**
|
|
31
|
+
3. Select **Claude Code Subscription**
|
|
32
|
+
|
|
33
|
+
## How It Works
|
|
34
|
+
|
|
35
|
+
When using Claude Code mode, Alucify runs AI analysis through the Claude CLI on your machine. This uses your subscription's included usage rather than per-token billing.
|
|
36
|
+
|
|
37
|
+
## Troubleshooting
|
|
38
|
+
|
|
39
|
+
If Alucify can't detect your Claude CLI:
|
|
40
|
+
|
|
41
|
+
- Ensure `claude` is available in your system PATH
|
|
42
|
+
- Try running `claude --version` in your terminal to verify it's installed
|
|
43
|
+
- Restart VS Code after installing the Claude CLI
|
|
@@ -0,0 +1,36 @@
|
|
|
1
|
+
---
|
|
2
|
+
order: 1
|
|
3
|
+
title: "Creating an Account & Signing In"
|
|
4
|
+
description: "How to create an Alucify account and sign in from the welcome screen."
|
|
5
|
+
---
|
|
6
|
+
|
|
7
|
+
# Creating an Account & Signing In
|
|
8
|
+
|
|
9
|
+
An Alucify account is required to use the extension. Accounts are free to create.
|
|
10
|
+
|
|
11
|
+
## Signing Up
|
|
12
|
+
|
|
13
|
+
1. Click the **Alucify** icon in the VS Code activity bar
|
|
14
|
+
2. On the welcome screen, find **Step 1: Sign in to Alucify**
|
|
15
|
+
3. Click **Sign Up / Sign In**
|
|
16
|
+
4. A browser window opens to the Alucify authentication page
|
|
17
|
+
5. Create a new account or sign in with an existing one
|
|
18
|
+
6. After successful authentication, VS Code shows a confirmation — click **Allow** to complete the sign-in
|
|
19
|
+
|
|
20
|
+
Once signed in, a green checkmark appears next to the step, along with your username.
|
|
21
|
+
|
|
22
|
+
## Signing Out
|
|
23
|
+
|
|
24
|
+
To sign out:
|
|
25
|
+
|
|
26
|
+
1. Open the Command Palette (`Ctrl+Shift+P` / `Cmd+Shift+P`)
|
|
27
|
+
2. Search for **Alucify: Logout**
|
|
28
|
+
3. Select the command
|
|
29
|
+
|
|
30
|
+
## Signing Back In
|
|
31
|
+
|
|
32
|
+
If you've previously dismissed the welcome screen:
|
|
33
|
+
|
|
34
|
+
1. Open the Command Palette
|
|
35
|
+
2. Search for **Alucify: Sign In**
|
|
36
|
+
3. Follow the browser authentication flow
|
|
@@ -0,0 +1,28 @@
|
|
|
1
|
+
---
|
|
2
|
+
order: 4
|
|
3
|
+
title: "Switching Between Auth Modes"
|
|
4
|
+
description: "How to switch between Anthropic API key and Claude Code subscription authentication."
|
|
5
|
+
---
|
|
6
|
+
|
|
7
|
+
# Switching Between Auth Modes
|
|
8
|
+
|
|
9
|
+
You can switch between API key and Claude Code subscription at any time.
|
|
10
|
+
|
|
11
|
+
## How to Switch
|
|
12
|
+
|
|
13
|
+
1. Open the Command Palette (`Ctrl+Shift+P` / `Cmd+Shift+P`)
|
|
14
|
+
2. Search for **Alucify: Switch Auth Mode**
|
|
15
|
+
3. Select your preferred mode
|
|
16
|
+
|
|
17
|
+
You'll see a confirmation message: "Auth mode switched to: [mode name]"
|
|
18
|
+
|
|
19
|
+
## When to Switch
|
|
20
|
+
|
|
21
|
+
- **To API Key** — If you don't have a Claude Code subscription, or if you want more control over per-token costs
|
|
22
|
+
- **To Claude Code** — If you have a Claude Code or Claude Max subscription and want to use your subscription's included usage
|
|
23
|
+
|
|
24
|
+
## Notes
|
|
25
|
+
|
|
26
|
+
- Switching auth modes does not affect your existing analysis data
|
|
27
|
+
- Any in-progress workflows will need to be restarted after switching
|
|
28
|
+
- Your API key (if previously set) is retained in secure storage even when using Claude Code mode, so you can switch back without re-entering it
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{ "title": "Chat", "order": 7 }
|
|
@@ -0,0 +1,39 @@
|
|
|
1
|
+
---
|
|
2
|
+
order: 1
|
|
3
|
+
title: "Asking Questions About Your Project"
|
|
4
|
+
description: "How to use the chat panel to ask questions about your project's architecture and specs."
|
|
5
|
+
---
|
|
6
|
+
|
|
7
|
+
# Asking Questions About Your Project
|
|
8
|
+
|
|
9
|
+
The Ask About Project feature lets you have a conversation with Claude about your project, using the context from your Alucify analysis.
|
|
10
|
+
|
|
11
|
+
## How to Start
|
|
12
|
+
|
|
13
|
+
1. In the **Project Analysis** panel header, click the **chat icon** (comment discussion icon)
|
|
14
|
+
2. Or open the Command Palette and search for **Alucify: Ask about Project**
|
|
15
|
+
|
|
16
|
+
A chat panel opens in the secondary sidebar.
|
|
17
|
+
|
|
18
|
+
## What You Can Ask
|
|
19
|
+
|
|
20
|
+
With your analysis as context, Claude can answer questions like:
|
|
21
|
+
|
|
22
|
+
- "What does the authentication flow look like?"
|
|
23
|
+
- "Which parts of the codebase handle payment processing?"
|
|
24
|
+
- "Are there any requirements that aren't implemented yet?"
|
|
25
|
+
- "What would be affected if I changed the user data model?"
|
|
26
|
+
- "How does the notification system work?"
|
|
27
|
+
|
|
28
|
+
## Using the Chat
|
|
29
|
+
|
|
30
|
+
- Type your question in the message input field
|
|
31
|
+
- Press Enter or click the send button
|
|
32
|
+
- Claude responds with answers grounded in your project analysis
|
|
33
|
+
- Continue the conversation with follow-up questions
|
|
34
|
+
|
|
35
|
+
## Tips
|
|
36
|
+
|
|
37
|
+
- **Run analysis first** — The chat is most useful after you've run code analysis and/or spec analysis, as Claude uses those results to ground its answers
|
|
38
|
+
- **Be specific** — The more specific your question, the more targeted the answer
|
|
39
|
+
- **Use for exploration** — This is a great way to understand an unfamiliar codebase or to explore how different parts of your project relate to each other
|
|
@@ -0,0 +1,53 @@
|
|
|
1
|
+
---
|
|
2
|
+
order: 3
|
|
3
|
+
title: "Generating Artifacts in Chat"
|
|
4
|
+
description: "Creating new spec artifacts with AI assistance inside VS Code."
|
|
5
|
+
---
|
|
6
|
+
|
|
7
|
+
# Generating Artifacts in Chat
|
|
8
|
+
|
|
9
|
+
Artifact generation lets you create new specs from templates with Claude's help. Useful when you know what you want but don't have a formal spec document yet.
|
|
10
|
+
|
|
11
|
+
## Prerequisites
|
|
12
|
+
|
|
13
|
+
- Authentication set up (see [Authentication](../authentication/signing-in.md))
|
|
14
|
+
- For Claude Code subscription users: Claude Code CLI installed
|
|
15
|
+
|
|
16
|
+
## Starting Generation
|
|
17
|
+
|
|
18
|
+
1. Open the Specs sidebar view
|
|
19
|
+
2. Click **Generate Spec**
|
|
20
|
+
3. Pick a template:
|
|
21
|
+
- **PRD** — full product requirements document
|
|
22
|
+
- **Architecture** — technical architecture doc
|
|
23
|
+
- **UX** — user experience flow / wireframe description
|
|
24
|
+
- **Epic / Story** — feature epic with stories and acceptance criteria
|
|
25
|
+
- **Quick Spec** — lightweight one-pager
|
|
26
|
+
- **Custom** — describe your own format
|
|
27
|
+
|
|
28
|
+
4. Describe what you want in a few sentences
|
|
29
|
+
|
|
30
|
+
Claude drafts a full artifact and streams it into the chat view. You iterate with follow-up messages ("make it shorter", "add mobile considerations", etc.) until you're happy.
|
|
31
|
+
|
|
32
|
+
## What Claude Sees
|
|
33
|
+
|
|
34
|
+
Each generation prompt includes:
|
|
35
|
+
|
|
36
|
+
- The template structure
|
|
37
|
+
- Your description
|
|
38
|
+
- Your current [AppGraph](../../concepts/appgraph.md) summary (so the new spec is consistent with existing ones)
|
|
39
|
+
- Relevant code excerpts if your description mentions specific features
|
|
40
|
+
|
|
41
|
+
## Saving the Artifact
|
|
42
|
+
|
|
43
|
+
When you're ready, click **Save as Artifact**. The generated content is written to `.alucify/artifacts/` with a filename you choose. It immediately appears in your Specs list with **Pending** status.
|
|
44
|
+
|
|
45
|
+
## Running Analysis
|
|
46
|
+
|
|
47
|
+
Once saved, you can click **Analyze** on the artifact to incorporate it into the project AppGraph — same flow as any manually-imported artifact.
|
|
48
|
+
|
|
49
|
+
## Good Practices
|
|
50
|
+
|
|
51
|
+
- **Start specific** — vague prompts produce generic specs. Say what this feature does, who uses it, and why.
|
|
52
|
+
- **Reference existing nodes** — mention specific requirements, components, or data models by name. Claude will connect the new spec to them.
|
|
53
|
+
- **Iterate, don't restart** — use follow-up messages rather than starting over. Claude keeps context and improves on previous drafts.
|