Update code reviews

TrafficGuard · Aug 8, 2024 · 198811c · 198811c
1 parent 6241354
commit 198811c
Show file tree

Hide file tree

Showing 24 changed files with 425 additions and 252 deletions.
diff --git a/README.md b/README.md
@@ -18,12 +18,11 @@
 
 ## The Nous Story
 
-Nous started from a simple goal: to harness AI's potential to **enhance real-world productivity**, born in DevOps and Platform Engineering space. We envisioned a tool that could:
+Nous started from a simple goal: to harness AI's potential to enhance real-world productivity, born in DevOps and Platform Engineering space. We envisioned a tool that could:
 
-- Automate standard and simple requests using natural language prompts.
-- Intelligently triage build failures, support requests and production incidents.
+- Automate various processes and support requests, and triage build failures.
 - Review code for compliance with standards and best practices.
-- Assist with large/complex refactorings, and more.
+- Assist with large refactorings, and more.
 
 At TrafficGuard we process billions of events a month for our global clients, [increasing their ad spend ROI](https://www.trafficguard.ai/case-studies?ref=gh) by protecting against bots and other forms of invalid traffic.
 Our SaaS on GCP comprises projects developed in TypeScript, Python, GoogleSQL, PHP and Terraform, deployed from GitLab. 
@@ -42,7 +41,7 @@ Key features include:
 
 - Advanced autonomous agents
   - Reasoning/planning inspired from Google's [Self-Discover](https://arxiv.org/abs/2402.03620) paper
-  - Memory and function history for complex, multi-step workflows
+  - Memory and function call history for complex, multi-step workflows
   - Adaptive iterative planning with hierarchical task decomposition
   - Two control-loop function calling options (LLM-independent):
     - Custom XML-based function calling
@@ -65,7 +64,7 @@ Key features include:
 - Observability with OpenTelemetry tracing
 - Code Editing Agent:
   - Auto-detection of project initialization, compile, test and lint
-  - Find the relevant files to edit
+  - Find the relevant files to edit and perform initial analysis
   - Code editing loop with compile, lint, test, fix (editing delegates to [Aider](https://aider.chat/))
     - Compile error analyser can search online, add additional files and packages
 - Software Engineer Agent:
@@ -79,7 +78,7 @@ Key features include:
 
 ## UI Examples
 
-[New Agent](#new-agent) | [Sample trace](#sample-trace) | [Human in the loop notification](#human-in-the-loop-notification) | [Resume error](#resume-error) | [List agents](#list-agents)
+[New Agent](#new-agent) | [Sample trace](#sample-trace) | [Human in the loop notification](#human-in-the-loop-notification) | [Resume error](#resume-error) | [List agents](#list-agents) | [Code review config](#code-review-configuration)
 
 ### New Agent
 
@@ -93,14 +92,18 @@ Key features include:
 
 <img src="https://public.trafficguard.ai/nous/feedback.png" width="702">
 
-### Resume error
+### Agent requested feedback
 
-![Resume error](https://public.trafficguard.ai/nous/error.png)
+![Feedback requested](https://public.trafficguard.ai/nous/agent-feedback.png)
 
 ### List agents
 
 ![List agents](https://public.trafficguard.ai/nous/list.png)
 
+### Code review configuration
+
+![Code review configuration](https://public.trafficguard.ai/nous/code-review.png)
+
 ## Code Examples
 
 ### Nous vs LangChain
@@ -147,18 +150,32 @@ console.log(result);
 
 #### Nous
 ```typescript
-import { initAgentContext, llms } from '#agent/context'
+import { llms } from '#agent/context'
 import { anthropicLLMs } from '#llms/anthropic'
 
-initAgentContext(anthropicLLMs());
-
 const prompt1 = (person: string) => `What is the city ${person} is from? Only respond with the name of the city.`;
 const prompt2 = (city: string, language: string) => `What country is the city ${city} in? Respond in ${language}.`;
 
-const city = await llms().easy.generateText(prompt1('Obama'));
-const result = await llms().easy.generateText(prompt2(city, 'German'));
+runAgentWorkflow({ llms: anthropicLLMs() }, async () => {
+  const city = await llms().easy.generateText(prompt1('Obama'));
+  const result = await llms().easy.generateText(prompt2(city, 'German'));
 
-console.log(result);
+  console.log(result);
+});
+```
+
+The Nous code also has the advantage of static typing with the prompt arguments, enabling you to refactor with ease.
+Using simple control flow allows easy debugging with breakpoints/logging.
+
+To run a fully autonomous agent:
+
+```typescript
+startAgent({
+  agentName: 'Create ollama',
+  initialPrompt: 'Research how to use ollama using node.js and create a new implementation under the llm folder. Look at a couple of the other files in that folder for the style which must be followed',
+  functions: [FileSystem, Perplexity, CodeEditinAgent],
+  llms,
+});
 ```
 
 The Nous code also has the advantage of static typing with the prompt arguments, enabling you to refactor with ease.
@@ -176,10 +193,10 @@ Visit our [documentation site](https://nous.trafficguard.ai/setup/) for the gett
 
 ## Contributing
 
-We welcome contributions to the project through [issues](https://github.com/TrafficGuard/nous/issues), [pull requests](https://github.com/TrafficGuard/nous/pulls)  or [discussions](https://github.com/TrafficGuard/nous/discussions)
+We warmly welcome contributions to the project through [issues](https://github.com/TrafficGuard/nous/issues), [pull requests](https://github.com/TrafficGuard/nous/pulls)  or [discussions](https://github.com/TrafficGuard/nous/discussions)
 
 Contributed by [TrafficGuard](https://www.trafficguard.ai) - Increasing the ROI on your ad spend.
 
-
+Reach out to us as nous@trafficguard.ai if you'd like support to ramp up as a contributor.
 
 
diff --git a/docs/docs/index.md b/docs/docs/index.md
@@ -11,9 +11,9 @@
 
 Nous started from a simple goal: to harness AI's potential to enhance real-world productivity, born in DevOps and Platform Engineering space. We envisioned a tool that could:
 
-- Automate standard processes and triage build failures, support requests and production incidents.
+- Automate various processes and support requests, and triage build failures.
 - Review code for compliance with standards and best practices.
-- Assist with large/complex refactorings, and more.
+- Assist with large refactorings, and more.
 
 At TrafficGuard we process billions of events a month for our global clients, [increasing their Ad spend ROI](https://www.trafficguard.ai/case-studies?ref=nous) from bots and other invalid traffic. 
 Our platform on Google Cloud comprises projects developed in TypeScript, Python, GoogleSQL, PHP and Terraform, deployed from GitLab.
@@ -32,7 +32,7 @@ Some of the key features include:
 
 - Advanced autonomous agents
     - Reasoning/planning inspired from Google's [Self-Discover](https://arxiv.org/abs/2402.03620) paper
-    - Memory and function history for complex workflows
+    - Memory and function call history for complex workflows
     - Iterative planning with hierarchical task decomposition
     - Two LLM-independent function calling options:
         - Custom XML-based function calling
@@ -55,7 +55,7 @@ Some of the key features include:
 - Observability with OpenTelemetry tracing
 - Code Editing Agent:
     - Auto-detection of project initialization, compile, test and lint
-    - Find the relevant files to edit
+    - Find the relevant files to edit and perform initial analysis
     - Code editing loop with compile, lint, test, fix (editing delegates to [Aider](https://aider.chat/))
         - Compile error analyser can search online, add additional files and packages
     - Review the changes with an additional code editing loop if required.
@@ -82,14 +82,18 @@ Some of the key features include:
 
 <img src="https://public.trafficguard.ai/nous/feedback.png" width="702">
 
-### Resume error
+### Agent requested feedback
 
-![Resume error](https://public.trafficguard.ai/nous/error.png)
+![Feedback requested](https://public.trafficguard.ai/nous/agent-feedback.png)
 
 ### List agents
 
 ![List agents](https://public.trafficguard.ai/nous/list.png)
 
+### Code review configuration
+
+![Code review configuration](https://public.trafficguard.ai/nous/code-review.png)
+
 ## Code Examples
 
 ### Nous vs LangChain
@@ -137,19 +141,33 @@ console.log(result);
 #### Nous
 ```typescript
 import { llms } from '#agent/context'
+import { anthropicLLMs } from '#llms/anthropic'
 
 const prompt1 = (person: string) => `What is the city ${person} is from? Only respond with the name of the city.`;
 const prompt2 = (city: string, language: string) => `What country is the city ${city} in? Respond in ${language}.`;
 
-const city = await llms().easy.generateText(prompt1('Obama'));
-const result = await llms().easy.generateText(prompt2(city, 'German'));
+runAgentWorkflow({ llms: anthropicLLMs() }, async () => {
+  const city = await llms().easy.generateText(prompt1('Obama'));
+  const result = await llms().easy.generateText(prompt2(city, 'German'));
 
-console.log(result);
+  console.log(result);
+});
 ```
 
 The Nous code also has the advantage of static typing with the prompt arguments, enabling you to refactor with ease.
 Using simple control flow allows easy debugging with breakpoints/logging.
 
+To run a fully autonomous agent:
+
+```typescript
+startAgent({
+  agentName: 'Create ollama',
+  initialPrompt: 'Research how to use ollama using node.js and create a new implementation under the llm folder. Look at a couple of the other files in that folder for the style which must be followed',
+  functions: [FileSystem, Perplexity, CodeEditinAgent],
+  llms,
+});
+```
+
 ### Automated LLM function schemas
 
 LLM function calling schema are automatically generated by having the `@func` decorator on class methods, avoiding the
@@ -176,5 +194,6 @@ export class Jira {
 
 ## Contributing 
 
-We welcome contributions to the project through [issues](https://github.com/TrafficGuard/nous/issues), [pull requests](https://github.com/TrafficGuard/nous/pulls)  or [discussions](https://github.com/TrafficGuard/nous/discussions)
+We warmly welcome contributions to the project through [issues](https://github.com/TrafficGuard/nous/issues), [pull requests](https://github.com/TrafficGuard/nous/pulls)  or [discussions](https://github.com/TrafficGuard/nous/discussions)
 
+Reach out to us as nous@trafficguard.ai if you'd like support to ramp up as a contributor.
diff --git a/docs/docs/roadmap.md b/docs/docs/roadmap.md
@@ -6,12 +6,14 @@ GitHub support
 
 Ollama support
 
+https://github.com/nextauthjs/next-auth/pull/9587
+
 Additional agents:
 
 - Slack chatbot agent
 - SWE-bench agent
-- Build failure investigation agent
-- Autonomous agent which generates Python executed by Pyodide for tool function calling
+- Add tools for GAIA agent
+- CI/CD failure investigation agent
 
 Improvements to the autonomous and code editing agents (a whole roadmap in itself)
 

diff --git a/frontend/src/app/code-review/code-review-edit.component.html b/frontend/src/app/code-review/code-review-edit.component.html
@@ -17,7 +17,7 @@
           <div formGroupName="file_extensions" class="col">
             <mat-form-field appearance="fill" class="full-width">
               <mat-label>Included File Extensions</mat-label>
-              <mat-chip-list #chipList>
+              <mat-chip-list #chipList required>
                 <mat-chip
                   *ngFor="let ext of editForm.get('file_extensions.include')?.value"
                   [removable]="true"
@@ -38,7 +38,7 @@
           <div formGroupName="requires" class="col">
             <mat-form-field appearance="fill" class="full-width">
               <mat-label>Required Text In Diff</mat-label>
-              <mat-chip-list #requiredChipList>
+              <mat-chip-list #requiredChipList required>
                 <mat-chip
                   *ngFor="let text of editForm.get('requires.text')?.value"
                   [removable]="true"

diff --git a/frontend/src/app/code-review/code-review-edit.component.ts b/frontend/src/app/code-review/code-review-edit.component.ts
@@ -44,8 +44,8 @@ export class CodeReviewEditComponent implements OnInit {
       requires: this.fb.group({
         text: [[], [Validators.required, this.arrayNotEmpty]],
       }),
-      tags: [[], this.arrayNotEmpty],
-      project_paths: [[], this.arrayNotEmpty],
+      tags: [[]],
+      project_paths: [[]],
       examples: this.fb.array([], [Validators.required, this.arrayNotEmpty]),
     });
   }
@@ -59,7 +59,26 @@ export class CodeReviewEditComponent implements OnInit {
     this.isLoading = true;
     this.codeReviewService.getCodeReviewConfig(this.configId!).subscribe(
       (response) => {
-        this.editForm.patchValue(response.data);
+        const data = response.data;
+        this.editForm.patchValue(data);
+
+        // Clear existing examples
+        while (this.examples.length !== 0) {
+          this.examples.removeAt(0);
+        }
+
+        // Add examples from the loaded data
+        if (data.examples && Array.isArray(data.examples)) {
+          data.examples.forEach((example: any) => {
+            this.examples.push(
+              this.fb.group({
+                code: [example.code, Validators.required],
+                review_comment: [example.review_comment, Validators.required],
+              })
+            );
+          });
+        }
+
         this.isLoading = false;
       },
       (error) => {

diff --git a/package-lock.json b/package-lock.json
diff --git a/package.json b/package.json
@@ -92,6 +92,7 @@
     "ignore": "^5.3.1",
     "jsdom": "^24.0.0",
     "lodash": "^4.17.20",
+    "micromatch": "^4.0.7",
     "module-alias": "^2.2.2",
     "openai": "^4.28.4",
     "pino": "^8.18.0",