Commit Graph
7 Commits
Author SHA1 Message Date
rootandClaude Opus 5.5 0508f3418f feat(agent): require Chinese thinking and reports in subagent prompts (AG-4)
Subagent / vision subflow prompts (audience=agent) get a dedicated section:
report goes to the parent session, and thinking, step notes and reports stay
in Chinese even when the goal, tool results or web pages are in English.
Prompt regression test covers both the shared hard rule and the new section.

Child reports reaching the user are already humanized: the child session page
uses the same FIX-7 report-humanize path, and the parent's final report resolves
child run IDs to their goal titles; agent.spawn/await tool rows show no raw summary.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-30 17:58:08 +08:00
rootandClaude Opus 5.5 2ab6d452dd feat(agent): 系统提示规定汇报格式(AGUI-3)
新增「汇报格式」一节:第一句写结论;对象用名字,不写函数名、工具名、字段名、UUID 或 hash;
不超过 10 行;列举 3 个以上资源才用表格;不写标题和 HTML。证据规则改为写对象名称和关键数值。
方案规则改为只在真有分歧时给方案,并说明方案绑定计划、选定即批准的用法;步骤标题不写工具名。
子会话与视觉子流程汇报给父会话(audience: agent),不加用户汇报格式。
加 prompt 回归测试,快照断言汇报格式的关键约束句。

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-29 22:41:13 +08:00
rootandClaude Opus 5.5 c437c6bff6 feat(browser): vision automation loop with screenshots, coordinate actions and bounded challenge solving
- browser.screenshot returns the page as an image content block (≤1280px wide, JPEG quality
  configurable) for vision-capable upstreams; non-vision upstreams get a redacted text snapshot.
  Login pages / password fields are marked sensitive and withheld unless a
  browser.sensitive_screenshot grant exists and allowSensitive is passed.
- browser.click_at / type_at / scroll / wait act in screenshot pixels; the server converts via
  viewport/DPR, requires automation control and an allowlisted host, and writes a thumbnail
  evidence file (DATA_DIR/tmp/evidence) referenced from agent_run_event.
- browser.solve_challenge runs a bounded child agent_run (vision caps only, maxSteps ≤ 12,
  budget), judges success from page facts, stops immediately on 2FA / SMS / device confirmation,
  and otherwise moves the session to waiting with evidence and a takeover link.
- PiAgentRuntime passes modelImage outputs as pi-ai image content when model.input has image;
  the ledger never stores image data; coordinate actions use per-invocation idempotency keys.
- Tests: image codec, runtime image content, E14 e2e (login solved, 2FA stop, step limit,
  sensitive withheld). Real camofox 1.17.0 check on example.com recorded in the runbook.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-29 01:27:48 +08:00
rootandClaude Opus 5.5 743fe2486d feat(agent): per-upstream pi Models with isolated credential stores
- PiUpstreamClient: one Models per upstream holding only its provider; credential
  store adapter maps to that upstream's secret, so several accounts of one provider
  coexist; host env is only read for authKind none (ambient cloud creds)
- provider catalog from builtinProviders() plus static oauth modes, env fields and
  subscription terms notes; openai-compatible custom endpoints
- fake upstream provider with mock OAuth (browser paste / device approval)
- PiAgentRuntime accepts an upstream; thinking level clamped to the model

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-28 02:23:09 +08:00
rootandClaude Fable 5.1 304feea141 chore(deno): migrate toolchain from Node 24 + pnpm to Deno 2.9.7 + TypeScript 7.0.2 (W1.5)
- root deno.json workspace (nodeModulesDir auto), npm versions registered once in imports, deno.lock;
  remove pnpm-lock/pnpm-workspace/root package.json; members keep minimal package.json (name/type/exports)
- tasks: dev, typecheck (tsc 7.0.2 per member + deno check), lint/fmt (biome, useImportExtensions),
  test (vitest via deno), check:tokens, build (web + deno compile), ci, db:migrate/reset, gen:types
- tsconfig base moved to config/ (no root tsconfig.json); deno types -> types/deno.d.ts; web drops baseUrl
- server: Deno.serve entry with graceful shutdown, --role migrate|preflight|probe, WEB_ROOT static SPA hosting;
  drop @hono/node-server, tsx, esbuild build.mjs
- logger: redact child() bindings and serialize err by message/stack; sync stdout destination; tests
- relative imports get explicit extensions (sites/mp-sync paths left for follow-up)
- crash-recovery tests spawn `deno run -A`; tasks tests use unique per-file schemas
- single-container Dockerfile + entrypoint (migrate -> preflight -> api+worker watchdog), docker-compose.yml
- docs: README, runbook, ADR-001 superseded, plan §1 validation row

Co-Authored-By: Claude Fable 5.1 <[email protected]>
2026-09-27 18:53:06 +08:00
rootandClaude Fable 5.1 9c881e737f feat(agent): capability registry, Pi AgentRuntimePort, fake model provider and redaction guard (W1-5)
- CapabilityRegistry: descriptor(id/version/schema/scope/risk/idempotent/preview/replay),
  invoke = scope limit -> input schema -> server-side authorize -> idempotency key -> handler
  -> output schema -> audit; capability id <-> tool name mapping (download.submit <-> download__submit)
- PiAgentRuntime (pi-agent-core AgentHarness 0.87.1): whitelist tools + before_tool re-check,
  redacted event stream, JSONL session at sessionDir/<runId>.jsonl, resume/abort/compact,
  tool-call/duration/token/cost budgets -> abort + run_failed(budget)
- AgentSettings -> pi-ai createProvider (anthropic / openai-compatible), compaction mapping
- FakeModelProvider + createFakeRuntime (scripted, transcript-derived, survives crash/resume)
- buildSystemPrompt, redactForModel; 27 tests incl. SIGKILL + resume

Co-Authored-By: Claude Fable 5.1 <[email protected]>
2026-09-27 17:55:28 +08:00
umtandClaude Fable 5.1 b22c440139 chore: scaffold monorepo and migrate uhd_web design system (W0-A)
- pnpm workspace: apps/{web,server}, packages/{contracts,db,domain,adapters,tasks,agent,testkit}
- server: Hono + zod-openapi, API Key auth with scopes, redacting pino logger, SSE event bus with replay, worker heartbeat
- db: Kysely + pg, file migrator, 0001_init (settings/api_key/audit_event/outbox_event)
- web: React Router 7 SPA console shell with 12 workspace placeholders, uhd_web tokens/ui/guard scripts
- docs: ADR-001 (Node over Deno), dev runbook, design system rules

Co-Authored-By: Claude Fable 5.1 <[email protected]>
2026-09-27 17:36:26 +08:00