@tea-agent/loop-agent 0.1.0-fe-test.0

package/docs/templates/frontend-test-dag.json

@@ -0,0 +1,409 @@
+{
+  "$schema": "./agent-dag.schema.json",
+  "version": 3,
+  "title": "Frontend test RAG DAG template",
+  "runtimeContract": {
+    "schemaVersion": 1,
+    "agentRuntime": "pi-only",
+    "repairWriterProtocol": "explicit-node-v1"
+  },
+  "objective": "Build frontend test RAG package, generate Markdown cases, checklist+manifest, execute serially via playwright-cli, finalize frontend-test-result-v1, and write main HTML report (pipeline acceptance).",
+  "globalConstraints": [
+    "Do not generate pytest or Playwright source code.",
+    "Only use declared isolated test environments; production URLs and real credentials are blocked.",
+    "Every generated browser start command uses playwright-cli open --browser=chrome followed by the concrete URL selected by retrieve-frontend-test-context-pi from bound task source/config and written to context.md; the environment shell must mark that URL reachable before generation/execution; executable case lines never retain an angle-bracket URL/ref placeholder; subsequent commands stay in that default session and must not use unverified named-session flags.",
+    "Case children execute serially. Persist each case result, logs and browser evidence before the next child starts.",
+    "A token threshold is a post-case stop check, not a model hard token cap; unstarted cases must be recorded as blocked: token-budget-exhausted.",
+    "Default pipeline acceptance is the final frontend-test-result-v1 plus testcase/frontend/reports/frontend-test-report.html",
+    "Default frontendTest.reviewMode=off uses mechanical checklist-shell before materialize; set reviewMode=blocking for legacy dual LLM review gate.",
+    "playwright-cli-only: generators and executors may call only skill-declared playwright-cli commands; bare playwright / npx playwright / @playwright/test / Playwright source are forbidden with no native Playwright fallback.",
+    "Browser-tool preflight (preflight-frontend-browser-tool-shell) must reject CODE_AGENT_PI_BACKEND=cli-only, verify the Pi SDK structured custom-tool surface, and confirm the verified playwright-cli launcher + --help open/close/find/snapshot/click contract before any frontend-test Pi node; it does not select or freeze the tested URL; missing capability/CLI fails with zero Pi calls.",
+    "Case executors use structured playwright_cli custom tool under commandPolicy capability-allowlist; playwright-cli stays capability-gated while ordinary writers have bash.",
+    "File outputs use canonical --filename: playwright-cli screenshot --filename final.png (a real target/ref may precede it), playwright-cli pdf --filename final.pdf, and playwright-cli snapshot --filename snapshot.txt only when a snapshot file is needed; a snapshot without filename is response-only. Never use --path, --output, --file, or an output path as a positional target.",
+    "Passed cases require same-child ordered controller receipts: successful open → successful find → successful post-execution cleanup. Pre-start cleanup, snapshot/goto/screenshot/request/console, and ordinary interactions cannot establish passed authority; model prose cannot fake green.",
+    "Environment preflight must parse the retrieve-authored context.md baseUrl, reject unsafe/non-production violations, and curl-probe it before generate; unreachable or curl-unavailable ends preflight as blocked (frontend-base-url-unreachable|curl-unavailable) so generate/map do not run.",
+    "U/D cases must prove current-user data ownership or create cleanable current-user data or authorized Mock; otherwise blocked (current-user-data-unavailable|data-ownership-unverifiable|safe-test-data-setup-unavailable) without cross-user mutation.",
+    "Rerun topology is bounded by frontendTest.maxRerunAttempts (0..4): lean off/advisory defaults to 1 round, full reviewMode=blocking defaults to 2 rounds, and an explicit value overrides either default. The runtime hybrid emits one select+map pair per attempt round; each round only candidates blocked or missing-result cases and rewrites the authoritative case-result.json/execution.md, so the final round's evidence controls the report."
+  ],
+  "tasks": [
+    {
+      "id": "preflight-frontend-browser-tool-shell",
+      "depends_on": [],
+      "executor": "shell",
+      "role": "verifier",
+      "complexity": "LOW",
+      "writePolicy": "read-only",
+      "allowedPaths": [],
+      "forbiddenPaths": [
+        ".harness/**",
+        "artifacts/**"
+      ],
+      "outputContract": "Deterministic SDK-only browser-tool capability preflight before any frontend-test Pi node; no target URL selection; fail closed with browser-command-capability-unavailable | playwright-cli-unavailable | playwright-cli-contract-incompatible.",
+      "subtask_prompt": "Reject cli-only rollback, verify the Pi SDK structured custom-tool capability, and verify the controller-resolved playwright-cli launcher plus --help lists open/close/find/snapshot/click. Do not resolve or freeze the tested URL. Do not install packages. Do not start a browser session.",
+      "shell": {
+        "commands": [],
+        "frontendBrowserToolPreflight": {},
+        "cwd": ".",
+        "timeoutMs": 60000
+      }
+    },
+    {
+      "id": "prepare-frontend-test-package-shell",
+      "depends_on": [
+        "preflight-frontend-browser-tool-shell"
+      ],
+      "executor": "shell",
+      "role": "verifier",
+      "complexity": "LOW",
+      "writePolicy": "exclusive",
+      "writeSet": [
+        "testcase/frontend/rag/**"
+      ],
+      "allowedPaths": [
+        "testcase/frontend/rag/**"
+      ],
+      "forbiddenPaths": [
+        ".harness/**",
+        "artifacts/**"
+      ],
+      "outputContract": "Write testcase/frontend/rag/standard-scenarios.v1.json for generate-time standard scenario coverage.",
+      "subtask_prompt": "Materialize frontend-test standard scenarios v1 into the RAG package.",
+      "shell": {
+        "commands": [
+          "node -e \"const fs=require('fs'),path=require('path');const dest='testcase/frontend/rag/standard-scenarios.v1.json';const candidates=[path.join('docs','templates','frontend-test-standard-scenarios.v1.json')];let src=null;for(const c of candidates){if(fs.existsSync(c)){src=c;break;}}fs.mkdirSync(path.dirname(dest),{recursive:true});if(src){fs.copyFileSync(src,dest);process.stdout.write(JSON.stringify({status:'copied',from:src,to:dest}));}else{const minimal={schemaVersion:1,id:'frontend-test-standard-scenarios-v1',scenarios:[{id:'STD-FE-SMOKE-ENTRY',title:'入口可打开',category:'smoke',priority:'must',testPoints:['open'],minCases:1}]};fs.writeFileSync(dest,JSON.stringify(minimal,null,2)+'\\n');process.stdout.write(JSON.stringify({status:'fallback',to:dest}));}"
+        ],
+        "cwd": ".",
+        "timeoutMs": 60000
+      }
+    },
+    {
+      "id": "retrieve-frontend-test-context-pi",
+      "depends_on": [
+        "prepare-frontend-test-package-shell"
+      ],
+      "executor": "pi",
+      "role": "planner",
+      "toolProfile": "write",
+      "complexity": "MED",
+      "writePolicy": "exclusive",
+      "writeSet": [
+        "testcase/frontend/rag/**"
+      ],
+      "allowedPaths": [
+        "REPLACE/WITH/SOURCE/PATH/**",
+        "testcase/frontend/rag/**"
+      ],
+      "forbiddenPaths": [
+        ".harness/**",
+        "artifacts/**"
+      ],
+      "outputContract": "RAG context.md and coverage-map.md with baseUrl, baseUrlSource, environmentProbe=pending.",
+      "subtask_prompt_markdown": "./frontend-test-dag.retrieve-context.prompt.md",
+      "writeGuardPolicy": "tools-only"
+    },
+    {
+      "id": "materialize-frontend-test-execution-shell",
+      "depends_on": [
+        "retrieve-frontend-test-context-pi"
+      ],
+      "executor": "shell",
+      "role": "verifier",
+      "complexity": "LOW",
+      "writePolicy": "exclusive",
+      "writeSet": [
+        "testcase/frontend/rag/**"
+      ],
+      "allowedPaths": [
+        "testcase/frontend/rag/**"
+      ],
+      "forbiddenPaths": [
+        ".harness/**",
+        "artifacts/**"
+      ],
+      "outputContract": "Fail-closed environment preflight: absolute non-production baseUrl + curl HTTP reachability; writes environmentProbe facts; unreachable => blockedReason frontend-base-url-unreachable (node ERROR so generate/map do not run).",
+      "subtask_prompt": "Parse the concrete baseUrl selected by retrieve-frontend-test-context-pi from testcase/frontend/rag/context.md. Reject production, non-http(s), credentials, query and fragment. Probe with curl (HEAD then GET fallback; connect/max-time; no auth/cookie). 2xx/3xx => reachable and continue. 4xx/5xx/DNS/timeout/connection refused/TLS => blockedReason frontend-base-url-unreachable. Missing curl => blockedReason curl-unavailable. Do not start the app. Runtime hybrid generator embeds the authoritative probe script.",
+      "shell": {
+        "commands": [
+          "node -e \"console.log('template placeholder: runtime hybrid DAG embeds curl preflight; do not use this static command as source of truth')\""
+        ],
+        "cwd": ".",
+        "timeoutMs": 60000
+      }
+    },
+    {
+      "id": "generate-frontend-functional-cases-pi",
+      "depends_on": [
+        "materialize-frontend-test-execution-shell"
+      ],
+      "executor": "pi",
+      "role": "implementer",
+      "toolProfile": "write",
+      "complexity": "HIGH",
+      "writePolicy": "exclusive",
+      "writeSet": [
+        "testcase/frontend/cases/FE-*.md",
+        "testcase/frontend/cases/index.md",
+        "testcase/frontend/cases/manifest.draft.json"
+      ],
+      "allowedPaths": [
+        "testcase/frontend/rag/**",
+        "testcase/frontend/cases/**"
+      ],
+      "forbiddenPaths": [
+        ".harness/**",
+        "artifacts/**"
+      ],
+      "outputContract": "Markdown cases, index.md and manifest.draft.json schemaVersion 1 only; the exclusive materializer promotes the validated draft to manifest.json, and this generator must not write manifest.json; no test source code.",
+      "subtask_prompt_markdown": "./frontend-test-dag.generate-cases.prompt.md",
+      "writeGuardPolicy": "tools-only"
+    },
+    {
+      "id": "checklist-and-materialize-manifest-shell",
+      "depends_on": [
+        "generate-frontend-functional-cases-pi"
+      ],
+      "role": "verifier",
+      "executor": "shell",
+      "complexity": "LOW",
+      "writePolicy": "exclusive",
+      "writeSet": [
+        "testcase/frontend/cases/**"
+      ],
+      "allowedPaths": [
+        "testcase/frontend/rag/**",
+        "testcase/frontend/cases/**"
+      ],
+      "forbiddenPaths": [
+        ".harness/**",
+        "src/**"
+      ],
+      "outputContract": "Mechanical checklist then atomic manifest.json materialization; stdout one final JSON line {cases}.",
+      "subtask_prompt": "Run deterministic checklist then materialize manifest.json from draft.",
+      "shell": {
+        "commands": [],
+        "frontendTestCaseManifest": {},
+        "cwd": ".",
+        "timeoutMs": 120000
+      }
+    },
+    {
+      "id": "execute-frontend-cases-map",
+      "depends_on": [
+        "checklist-and-materialize-manifest-shell"
+      ],
+      "executor": "static",
+      "role": "verifier",
+      "complexity": "LOW",
+      "writePolicy": "none",
+      "allowedPaths": [],
+      "forbiddenPaths": [
+        ".harness/**",
+        "artifacts/**"
+      ],
+      "outputContract": "Serial aggregate of browser case results.",
+      "subtask_prompt": "Expand the validated manifest into serial browser case children.",
+      "static": {
+        "resultMarkdown": "Frontend case map expansion barrier."
+      },
+      "dynamicExpansion": {
+        "type": "map_agent",
+        "workflowNodeId": "execute-frontend-cases-map",
+        "itemsFrom": "$.nodes['checklist-and-materialize-manifest-shell'].output.cases",
+        "itemName": "case",
+        "maxItems": 20,
+        "maxExpandedNodes": 20,
+        "childIdPrefix": "execute-frontend-case",
+        "tokenBudget": {
+          "maxTokensPerCase": 20000,
+          "maxTotalTokens": 200000
+        },
+        "childTask": {
+          "executor": "pi",
+          "role": "implementer",
+          "skills": [
+            "playwright-cli"
+          ],
+          "toolProfile": "write",
+          "commandPolicy": {
+            "mode": "capability-allowlist",
+            "capabilities": [
+              "playwright-cli"
+            ]
+          },
+          "complexity": "MED",
+          "subtaskPromptTemplate": "Primary job: EXECUTE {{case.caseId}} from {{case.casePath}} with skill playwright-cli (fresh Pi session). Use the structured playwright_cli tool for every browser action. Do not request or search for bash. Translate each playwright-cli line in the case Markdown into one playwright_cli tool call. The controller-owned browser capability freezes baseUrl from allowlisted hash-bound task reference or the localhost default when no URL candidate exists; context/case prose may reference but cannot establish or override that origin. Start only with playwright_cli command=open and controller-injected concrete frozen URL args (default session; no -s=). Dynamic refs: literal eX/eY are documentation placeholders, not tool args. Immediately before every structured playwright_cli call that references an element, parse the actual eNN from the immediately preceding latest snapshot and pass only that actual eNN. Never send literal eX/eY, and never reuse a stale ref after a new snapshot. File outputs are canonical: screenshot uses [--filename, final.png] (or [e5, --filename, final.png] only for a real target), pdf uses [--filename, final.pdf], and snapshot writes a file only with [--filename, snapshot.txt]; snapshot without filename is response-only. Never use --path, --output, --file, or an output path as a positional target. Passed authority requires same-child ordered controller receipts: successful open → successful find → successful post-execution cleanup. Only successful find is meaningful; snapshot, goto, screenshot, request/console, and ordinary interactions cannot establish passed. Only when preflight or playwright_cli tool explicitly fails may you write blocked evidence. Always persist {{case.evidenceDir}}execution.md and {{case.evidenceDir}}case-result.json (caseId, status passed|failed|blocked, evidencePaths; blocked needs blockedReason). Business failed/blocked is not a node failure. Close via playwright_cli command=close. Return compact JSON (<=1200 chars): {caseId,status,evidencePaths,errorSummary,tokens}.",
+          "outputContract": "Compact JSON <=1200 chars. Browser actions must use structured playwright_cli tool.",
+          "writePolicy": "exclusive",
+          "allowedPaths": [
+            "testcase/frontend/cases/{{case.caseId}}.md",
+            "testcase/frontend/rag/context.md",
+            "testcase/frontend/rag/coverage-map.md",
+            "testcase/frontend/evidence/{{case.caseId}}/**"
+          ],
+          "forbiddenPaths": [
+            ".harness/**",
+            "artifacts/**"
+          ],
+          "writeSet": [
+            "testcase/frontend/evidence/{{case.caseId}}/**"
+          ],
+          "writeGuardPolicy": "tools-only"
+        }
+      }
+    },
+    {
+      "id": "select-frontend-rerun-candidates-shell",
+      "depends_on": [
+        "execute-frontend-cases-map"
+      ],
+      "executor": "shell",
+      "role": "verifier",
+      "complexity": "LOW",
+      "writePolicy": "exclusive",
+      "writeSet": [
+        "testcase/frontend/evidence/**"
+      ],
+      "allowedPaths": [
+        "testcase/frontend/cases/**",
+        "testcase/frontend/evidence/**"
+      ],
+      "forbiddenPaths": [
+        ".harness/**",
+        "artifacts/**"
+      ],
+      "outputContract": "Round 1 of the bounded rerun topology (lean default 1 round; full reviewMode=blocking default 2): stdout final JSON line {cases:[...]} for blocked or missing-result-file cases with rerunAttempt < 1; the runtime hybrid generates one selector/map pair per configured maxRerunAttempts so the latest rerun evidence is authoritative.",
+      "subtask_prompt": "Select frontend-test cases eligible for bounded rerun.",
+      "shell": {
+        "commands": [
+          "node -e \"const fs=require('fs'),path=require('path');const manifestPath='testcase/frontend/cases/manifest.json';if(!fs.existsSync(manifestPath)){process.stdout.write(JSON.stringify({cases:[]}));process.exit(0);}const manifest=JSON.parse(fs.readFileSync(manifestPath,'utf8'));const cases=[];for(const c of (manifest.cases||[])){const evidenceDir=(c.evidenceDir||('testcase/frontend/evidence/'+c.caseId+'/')).replace(/\\/+$/,'')+'/';const resultPath=path.join(evidenceDir,'case-result.json');const execPath=path.join(evidenceDir,'execution.md');let reason=null;let attempt=0;let missing=false;if(!fs.existsSync(resultPath)){missing=true;reason='missing-result-files';}else{try{const r=JSON.parse(fs.readFileSync(resultPath,'utf8'));attempt=Number(r.rerunAttempt||0)||0;if(r.status==='blocked')reason='blocked';if(!r.status){missing=true;reason='missing-result-files';}}catch(e){missing=true;reason='missing-result-files';}}if(!fs.existsSync(execPath)){const st=reason;if(st!=='blocked'){missing=true;reason=reason||'missing-result-files';}}const should=(reason==='blocked'||missing)&&attempt<1;if(should){cases.push({caseId:c.caseId,casePath:c.casePath||('testcase/frontend/cases/'+c.caseId+'.md'),evidenceDir,dimension:c.dimension||'core',acIds:c.acIds||[],rerunAttempt:attempt+1,reason:reason||'blocked'});}}const out={schemaVersion:1,cases};fs.mkdirSync('testcase/frontend/evidence',{recursive:true});fs.writeFileSync('testcase/frontend/evidence/rerun-candidates.json',JSON.stringify(out,null,2)+'\\n');process.stdout.write(JSON.stringify({cases}));"
+        ],
+        "cwd": ".",
+        "timeoutMs": 120000
+      }
+    },
+    {
+      "id": "rerun-frontend-cases-map",
+      "depends_on": [
+        "select-frontend-rerun-candidates-shell"
+      ],
+      "executor": "static",
+      "role": "verifier",
+      "complexity": "LOW",
+      "writePolicy": "none",
+      "allowedPaths": [],
+      "forbiddenPaths": [
+        ".harness/**",
+        "artifacts/**"
+      ],
+      "outputContract": "Serial rerun of blocked/missing-result frontend cases; rewrites authoritative case-result.json/execution.md so the final round's result controls reports.",
+      "subtask_prompt": "Expand rerun candidates into serial browser case children.",
+      "static": {
+        "resultMarkdown": "Frontend case map expansion barrier."
+      },
+      "dynamicExpansion": {
+        "type": "map_agent",
+        "workflowNodeId": "rerun-frontend-cases-map",
+        "itemsFrom": "$.nodes['select-frontend-rerun-candidates-shell'].output.cases",
+        "itemName": "case",
+        "maxItems": 20,
+        "maxExpandedNodes": 20,
+        "childIdPrefix": "rerun-frontend-case",
+        "tokenBudget": {
+          "maxTokensPerCase": 20000,
+          "maxTotalTokens": 200000
+        },
+        "childTask": {
+          "executor": "pi",
+          "role": "implementer",
+          "skills": [
+            "playwright-cli"
+          ],
+          "toolProfile": "write",
+          "commandPolicy": {
+            "mode": "capability-allowlist",
+            "capabilities": [
+              "playwright-cli"
+            ]
+          },
+          "complexity": "MED",
+          "subtaskPromptTemplate": "RERUN attempt {{case.rerunAttempt}} for {{case.caseId}} (reason={{case.reason}}). Rewrite authoritative evidenceDir case-result.json and execution.md; set rerunAttempt={{case.rerunAttempt}}; the final rewritten status is authoritative for reporting. Use fixed execution.md headings 执行摘要 and 实际执行步骤 when available. Primary job: EXECUTE {{case.caseId}} from {{case.casePath}} with skill playwright-cli (fresh Pi session). Use the structured playwright_cli tool for every browser action. Do not request or search for bash. Translate each playwright-cli line in the case Markdown into one playwright_cli tool call. The controller-owned browser capability freezes baseUrl from allowlisted hash-bound task reference or the localhost default when no URL candidate exists; context/case prose may reference but cannot establish or override that origin. Start only with playwright_cli command=open and controller-injected concrete frozen URL args (default session; no -s=). Dynamic refs: literal eX/eY are documentation placeholders, not tool args. Immediately before every structured playwright_cli call that references an element, parse the actual eNN from the immediately preceding latest snapshot and pass only that actual eNN. Never send literal eX/eY, and never reuse a stale ref after a new snapshot. File outputs are canonical: screenshot uses [--filename, final.png] (or [e5, --filename, final.png] only for a real target), pdf uses [--filename, final.pdf], and snapshot writes a file only with [--filename, snapshot.txt]; snapshot without filename is response-only. Never use --path, --output, --file, or an output path as a positional target. Passed authority requires same-child ordered controller receipts: successful open → successful find → successful post-execution cleanup. Only successful find is meaningful; snapshot, goto, screenshot, request/console, and ordinary interactions cannot establish passed. Only when preflight or playwright_cli tool explicitly fails may you write blocked evidence. Always persist {{case.evidenceDir}}execution.md and {{case.evidenceDir}}case-result.json (caseId, status passed|failed|blocked, evidencePaths; blocked needs blockedReason). Business failed/blocked is not a node failure. Close via playwright_cli command=close. Return compact JSON (<=1200 chars): {caseId,status,evidencePaths,errorSummary,tokens}.",
+          "outputContract": "Compact JSON <=1200 chars. Browser actions must use structured playwright_cli tool.",
+          "writePolicy": "exclusive",
+          "allowedPaths": [
+            "testcase/frontend/cases/{{case.caseId}}.md",
+            "testcase/frontend/rag/context.md",
+            "testcase/frontend/rag/coverage-map.md",
+            "testcase/frontend/evidence/{{case.caseId}}/**"
+          ],
+          "forbiddenPaths": [
+            ".harness/**",
+            "artifacts/**"
+          ],
+          "writeSet": [
+            "testcase/frontend/evidence/{{case.caseId}}/**"
+          ],
+          "writeGuardPolicy": "tools-only"
+        }
+      }
+    },
+    {
+      "id": "finalize-frontend-test-result-shell",
+      "depends_on": [
+        "rerun-frontend-cases-map"
+      ],
+      "role": "verifier",
+      "executor": "shell",
+      "complexity": "LOW",
+      "writePolicy": "exclusive",
+      "writeSet": [
+        "testcase/frontend/evidence/**"
+      ],
+      "allowedPaths": [
+        "testcase/frontend/cases/**",
+        "testcase/frontend/evidence/**"
+      ],
+      "forbiddenPaths": [
+        ".harness/**",
+        "src/**"
+      ],
+      "outputContract": "Evidence validation (advisory missing/malformed does not fail the node; hard-fail only path escape) then hash-bound frontend-test-result-v1. Node success means result-v1 was written, not that all cases passed.",
+      "subtask_prompt": "Validate case evidence then materialize authoritative frontend-test-result-v1. Missing/malformed case evidence is advisory; only unsafe evidence paths hard-fail.",
+      "shell": {
+        "commands": [],
+        "frontendTestResultFinalize": {},
+        "cwd": ".",
+        "timeoutMs": 120000
+      }
+    },
+    {
+      "id": "frontend-test-reports-shell",
+      "depends_on": [
+        "finalize-frontend-test-result-shell"
+      ],
+      "role": "verifier",
+      "executor": "shell",
+      "complexity": "LOW",
+      "writePolicy": "exclusive",
+      "writeSet": [
+        "testcase/frontend/reports/**"
+      ],
+      "allowedPaths": [
+        "testcase/frontend/**"
+      ],
+      "forbiddenPaths": [
+        ".harness/**",
+        "src/**"
+      ],
+      "outputContract": "Main frontend-test-report.md/html from frontend-test-result-v1; optional L-5. Pipeline acceptance = result-v1 + main HTML.",
+      "subtask_prompt": "Render operational reports from frontend-test-result-v1 only.",
+      "shell": {
+        "commands": [],
+        "frontendTestReports": {
+          "l5": true
+        },
+        "cwd": ".",
+        "timeoutMs": 120000
+      }
+    }
+  ]
+}

package/docs/templates/frontend-test-dag.retrieve-context.prompt.md

@@ -0,0 +1,21 @@
+# Retrieve frontend test context
+
+Write only `testcase/frontend/rag/context.md` and `coverage-map.md`. Record traceable facts from the task source, routes, components, API/Mock contracts, existing tests and execution contract. Do not invent fields, credentials, limits or test data.
+
+## Runtime Base URL resolution (required)
+
+Read the bound task source and environment/config references, then select one concrete browser entry URL. Prefer `TARGET_URL`/`targetUrl`, then `BASE_URL`/`baseUrl` or `FRONTEND_BASE_URL`/`frontendUrl`, then `LOGIN_URL`/`loginUrl`; accept common case, underscore, kebab and space variants. If no usable URL exists, use `http://localhost:5173`.
+
+- Write the selected value and its bound source into `context.md` as `baseUrl: <url>` and `baseUrlSource: <source>|default-localhost-5173`.
+- Never select API-only endpoints, production hosts, credential URLs, query-bearing URLs or fragment URLs.
+- Write `environmentProbe: pending`. The following environment shell deterministically parses `context.md`, validates URL safety, probes curl HEAD→GET, and replaces this with `reachable`, `unreachable`, or `curl-unavailable` plus a structured `blockedReason`.
+- Include the exact browser start prefix using the selected concrete URL:
+  `playwright-cli open --browser=chrome <resolved-base-url>`.
+
+Do not claim the environment is reachable until preflight completes. Preflight does not start the application.
+
+
+## Standard scenario coverage
+
+- Copy or reference `docs/templates/frontend-test-standard-scenarios.v1.json` into `testcase/frontend/rag/standard-scenarios.v1.json` when available.
+- Add `## Standard scenario coverage` to coverage-map.md with planned/n/a for each must scenario.

package/docs/templates/frontend-test-dag.retrospect.prompt.md

@@ -0,0 +1,5 @@
+# Frontend test retrospective
+
+Write a dated report under `testcase/frontend/reports/` (for example `frontend-test-retrospect-<date>.md`) after `frontend-test-result-v1` materialization. Do **not** require outcome=pass; failed, incomplete, and blocked runs still need a report.
+
+Combine a short AC → case → browser-evidence review with the closeout: case coverage, passed/failed/blocked results (including `token-budget-exhausted`), evidence gaps, browser anomalies, residual risks, and an A/B/C/D maturity rating. Cite the deterministic case-evidence validation outcome. Passed cases should reference the successful `find` assertion plus screenshot or equivalent browser evidence and the ordered post-execution cleanup receipt when available; failed/blocked cases need explicit reasons. Blocked cases never count as passed; a missing or malformed `execution.md` / `case-result.json` is a verification gap, not a pass. Do not write under `docs/**`.

package/docs/templates/frontend-test-dag.review-cases.prompt.md

@@ -0,0 +1,5 @@
+# Review frontend cases
+
+Read the RAG files and Markdown cases only. First line must be `VERDICT: pass` or `VERDICT: request-revision`. Report AC coverage, case independence, evidence completeness, unsafe environment/data dependencies, and manifest issues. This verdict is a deterministic safety gate: `request-revision` blocks manifest materialization and browser execution.
+
+Every case must retain the exact browser-start command prefix with the resolved absolute baseUrl from `testcase/frontend/rag/context.md` (prefer task source `config.md`, else `http://localhost:5173`): `playwright-cli open --browser=chrome <resolved-base-url>`; session flags must not precede `open`, and subsequent commands must remain in its default session without `-s=` or assumed named-session binding. Verify every executable sub-scenario specifies fixture/reset, UI reset, fresh snapshot before element refs, a successful `playwright-cli find ...` semantic assertion after `open`, and an evidence write point. `snapshot`, `goto`, `screenshot`, `request`/`console`, and ordinary interactions cannot independently authorize passed. Verify each case requires both `execution.md` and `case-result.json` under its own evidence directory. The JSON result must contain matching `caseId`, `status` (`passed`, `failed`, or `blocked`) and `evidencePaths`; blocked cases must name a non-empty `blockedReason` and cannot count as passed.

package/docs/templates/frontend-test-dag.review-execution.prompt.md

@@ -0,0 +1,3 @@
+# Review frontend execution evidence
+
+Review AC → case → browser-evidence traceability only after deterministic evidence validation. Each manifest case must have `execution.md` and a valid `case-result.json` with matching `caseId`, `status`, and `evidencePaths`; `blocked` requires a non-empty `blockedReason`. A passed case additionally needs a successful `find` semantic assertion, screenshot or equivalent browser evidence, and same-child ordered controller-owned `playwright-cli` receipts: successful `open` → successful `find` → successful `post-execution` cleanup. `pre-start` cleanup, snapshot/goto/screenshot/request/console, and ordinary interactions cannot satisfy this gate. Failed and blocked cases need an explicit cause. Treat `token-budget-exhausted` and `browser-command-evidence-missing` as blocked; do not substitute model conclusions or static checks for browser evidence.

package/docs/templates/frontend-test-standard-scenarios.v1.json

@@ -0,0 +1,159 @@
+{
+  "schemaVersion": 1,
+  "id": "frontend-test-standard-scenarios-v1",
+  "scenarios": [
+    {
+      "id": "STD-FE-SMOKE-ENTRY",
+      "title": "入口可打开与壳层可见",
+      "category": "smoke",
+      "priority": "must",
+      "dimensionHint": "core",
+      "testPoints": ["冻结 baseUrl 可 open", "关键导航或页面壳可见"],
+      "minCases": 1
+    },
+    {
+      "id": "STD-FE-EMPTY-STATE",
+      "title": "列表空态",
+      "category": "empty",
+      "priority": "must",
+      "dimensionHint": "boundary",
+      "testPoints": ["无数据时空态文案或引导", "不白屏"],
+      "minCases": 1
+    },
+    {
+      "id": "STD-FE-LIST-PAGE",
+      "title": "列表渲染",
+      "category": "list",
+      "priority": "must",
+      "dimensionHint": "core",
+      "testPoints": ["表头/列或卡片列表可见", "有数据时行可渲染"],
+      "minCases": 1
+    },
+    {
+      "id": "STD-FE-PAGINATION",
+      "title": "分页或加载更多",
+      "category": "pagination",
+      "priority": "should",
+      "dimensionHint": "boundary",
+      "testPoints": ["翻页或加载更多可观察", "数据不足可 blocked 并写原因"],
+      "minCases": 0
+    },
+    {
+      "id": "STD-FE-CREATE-FORM-VALIDATION",
+      "title": "创建表单校验",
+      "category": "form",
+      "priority": "must",
+      "dimensionHint": "boundary",
+      "testPoints": ["必填/格式校验", "不提交明显脏数据"],
+      "minCases": 1
+    },
+    {
+      "id": "STD-FE-CREATE-SUCCESS",
+      "title": "创建成功反馈",
+      "category": "create",
+      "priority": "must",
+      "dimensionHint": "flow",
+      "testPoints": ["成功提示", "列表或详情可见新数据或等价反馈"],
+      "minCases": 1
+    },
+    {
+      "id": "STD-FE-VIEW-DETAIL",
+      "title": "详情查看",
+      "category": "detail",
+      "priority": "must",
+      "dimensionHint": "core",
+      "testPoints": ["从列表进入详情", "关键字段可读"],
+      "minCases": 1
+    },
+    {
+      "id": "STD-FE-UPDATE-OWNED",
+      "title": "更新归属数据",
+      "category": "update",
+      "priority": "should",
+      "dimensionHint": "flow",
+      "testPoints": ["仅可证明归属数据", "否则 blocked"],
+      "minCases": 0
+    },
+    {
+      "id": "STD-FE-DELETE-CONFIRM",
+      "title": "删除确认",
+      "category": "delete",
+      "priority": "should",
+      "dimensionHint": "flow",
+      "testPoints": ["确认框或等价保护", "归属规则"],
+      "minCases": 0
+    },
+    {
+      "id": "STD-FE-SEARCH-FILTER",
+      "title": "搜索筛选",
+      "category": "filter",
+      "priority": "must",
+      "dimensionHint": "core",
+      "testPoints": ["关键字或状态筛选结果变化"],
+      "minCases": 1
+    },
+    {
+      "id": "STD-FE-API-ERROR-TOAST",
+      "title": "后端错误可见反馈",
+      "category": "error",
+      "priority": "should",
+      "dimensionHint": "backend",
+      "testPoints": ["错误提示可观察", "可用 Mock"],
+      "minCases": 0
+    },
+    {
+      "id": "STD-FE-LOADING-DISABLED",
+      "title": "提交中防重复",
+      "category": "interaction",
+      "priority": "must",
+      "dimensionHint": "ux",
+      "testPoints": ["loading 或按钮禁用", "重复点击不重复提交（能观察才断言）"],
+      "minCases": 1
+    },
+    {
+      "id": "STD-FE-LOADING-EMPTY",
+      "title": "首屏或列表 loading 与空态衔接",
+      "category": "interaction",
+      "priority": "should",
+      "dimensionHint": "ux",
+      "testPoints": ["有 loading 指示或骨架", "结束后列表或空态，不长期白屏"],
+      "minCases": 0
+    },
+    {
+      "id": "STD-FE-FEEDBACK-TOAST",
+      "title": "操作反馈可见",
+      "category": "interaction",
+      "priority": "must",
+      "dimensionHint": "ux",
+      "testPoints": ["成功或失败 toast 或行内提示可 find"],
+      "minCases": 1
+    },
+    {
+      "id": "STD-FE-KEYBOARD-FOCUS",
+      "title": "关键表单键盘与焦点",
+      "category": "interaction",
+      "priority": "should",
+      "dimensionHint": "ux",
+      "testPoints": ["Tab 或 Enter 或可见焦点", "无键盘信息则 GAP"],
+      "minCases": 0
+    },
+    {
+      "id": "STD-FE-DESTROY-RESET",
+      "title": "离开或取消后表单不脏留",
+      "category": "interaction",
+      "priority": "should",
+      "dimensionHint": "ux",
+      "testPoints": ["取消或返回后再进入，脏值已清或明确保留"],
+      "minCases": 0
+    },
+    {
+      "id": "STD-FE-OVERLAY-DISMISS",
+      "title": "弹层可关闭",
+      "category": "interaction",
+      "priority": "should",
+      "dimensionHint": "ux",
+      "testPoints": ["对话框或抽屉关闭后主界面可 find"],
+      "minCases": 0
+    }
+  ]
+}