feat: de-risk agent loop — semantic tool schema + agent loop + real-WP spike harness
Mirror to GitHub / mirror (push) Canceled after 0s
Mirror to GitHub / mirror (push) Canceled after 0s
- api/agents: tool-schemas (semantic allowlist), agent-loop (multi-step + budget), circuit-breaker, llm-client/tool-executor interfaces - api/agents: WpRestExecutor (real WP REST), OpenRouterLlmClient - e2e/agent: multi-step prompts + run-agent-spike runner (scores rendered changes) - ADR 0017; 101 unit tests green
This commit is contained in:
+2
-1
@@ -5,7 +5,8 @@
|
||||
"scripts": {
|
||||
"test": "vitest run",
|
||||
"golden": "tsx golden/src/run-golden.ts",
|
||||
"mirror:time": "tsx golden/src/run-mirror-timing.ts"
|
||||
"mirror:time": "tsx golden/src/run-mirror-timing.ts",
|
||||
"agent:spike": "tsx agent/run-agent-spike.ts"
|
||||
},
|
||||
"devDependencies": {
|
||||
"@playwright/test": "^1.62.1",
|
||||
|
||||
Reference in New Issue
Block a user