feat: inject past failures into agent prompts

Fetch relevant failures from API before spawning agent.
Failures appear as "Past Failures (avoid repeating these)" section
in the agent's system prompt.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Author: recuu-pfeg

src/api-client.ts

@@ -101,6 +101,7 @@ export interface ApiClient {
   fetchApiDocs(agentName: string): Promise<string>;
   fetchAgentMemories(agentName: string): Promise<AgentMemory[]>;
   putAgentMemory(agentName: string, key: string, data: { type: string; content: string }): Promise<void>;
+  fetchRelevantFailures(): Promise<Array<{ summary: string; failure_type: string; agent_name: string | null; created_at: string }>>;
 }
 
 export function createApiClient(apiUrl: string, apiKey: string): ApiClient {
@@ -319,5 +320,15 @@ export function createApiClient(apiUrl: string, apiKey: string): ApiClient {
         // Non-fatal
       }
     },
+
+    async fetchRelevantFailures(): Promise<Array<{ summary: string; failure_type: string; agent_name: string | null; created_at: string }>> {
+      try {
+        const res = await fetch(`${apiUrl}/api/v1/failures/relevant?limit=5`, { headers });
+        if (!res.ok) return [];
+        return (await res.json()) as Array<{ summary: string; failure_type: string; agent_name: string | null; created_at: string }>;
+      } catch {
+        return [];
+      }
+    },
   };
 }

src/cli.ts

@@ -234,6 +234,10 @@ async function runLoop(cliArgs: CliArgs, runner: AgentRunner): Promise<void> {
       const contextNotes = (task as Record<string, unknown>).context_notes as string | undefined;
       const fullDescription = [task.description, contextNotes].filter(Boolean).join("\n\n") || undefined;
 
+      // Fetch past failures for prompt injection
+      let pastFailures: Array<{ summary: string; failure_type: string; agent_name: string | null }> = [];
+      try { pastFailures = await api.fetchRelevantFailures(); } catch { /* non-fatal */ }
+
       const prompt = buildAgentPrompt({
         role: agentName, projectName: ctx.workspaceName, projectSpec: ctx.workspaceSpec,
         taskId: task.id, taskTitle: task.title,
@@ -244,6 +248,7 @@ async function runLoop(cliArgs: CliArgs, runner: AgentRunner): Promise<void> {
         playbookRules: (await ctx.api.fetchPlaybookPrompt(agentName)) || ctx.playbookRules,
         targetRepo: task.target_repo ?? undefined,
         apiDocs: apiDocs || undefined, engineHint: getEngine(cliArgs.engine).promptHint,
+        pastFailures: pastFailures.length > 0 ? pastFailures : undefined,
       });
 
       try {

src/prompt.ts

@@ -44,6 +44,8 @@ export interface PromptContext {
   apiDocs?: string;
   /** Engine-specific prompt hint (e.g. "CLAUDE.md is auto-loaded") */
   engineHint?: string;
+  /** Past failures relevant to this task (from Failure Database) */
+  pastFailures?: Array<{ summary: string; failure_type: string; agent_name: string | null }>;
 }
 
 const ROLE_DESCRIPTIONS: Record<string, string> = {
@@ -183,9 +185,13 @@ export function buildAgentPrompt(ctx: PromptContext): string {
 
   const engineHintLine = ctx.engineHint ? `\n${ctx.engineHint}` : "";
 
+  const failuresBlock = ctx.pastFailures?.length
+    ? `\n\n## Past Failures (avoid repeating these)\n${ctx.pastFailures.map((f) => `- [${f.failure_type}] ${f.summary}`).join("\n")}\n`
+    : "";
+
   return `${roleDesc}${langLine}${projectLine}${specBlock}
 ${securityRules}${playbookBlock}${repoBlock}${modeHeader}${extraRules}${engineHintLine}
-Your task: ${ctx.taskTitle}${priorityLine}${typeLine}${targetRepoLine}${descriptionBlock}
+Your task: ${ctx.taskTitle}${priorityLine}${typeLine}${targetRepoLine}${descriptionBlock}${failuresBlock}
 ${apiDocsBlock}
 ${completionInstructions}