[rig-summarizer] Daily file summary — 2026-10-11
Los mantenedores suelen responder en 1 día
Nadie ha tomado este issue todavía.
Evaluación
- Dificultad
- 5/5
- Tiempo estimado
- Más de una semana
- Aptitud para principiantes
- 1/100
- Tipo de issue
- Nueva funcionalidad
- Claridad
- Necesita aclaración
- Estado de actividad
- Activo
- Stack tecnológico
- github, typescript
- Área
- tooling
Línea de trabajo
The issue contains a generated summary of 20 TypeScript files, but it does not describe a requested change or a way to tell when work is complete. Read .agents/skills/rig/rig.ts and the listed engine, fixture, and integration test files to understand the project; the issue provides no implementation target or acceptance criteria.
Escrito por el modelo de indexación a partir del texto del issue.
Descripción
Global Summary
This workspace is the rig TypeScript agent harness: a single ~3,500-line core runtime (skills/rig/rig.ts) exposes declarative schema builders, prompt intents, and sub-agent/workflow orchestration on top of a Copilot SDK-backed engine, with optional Anthropic, Codex, DeepSeek, Gemini, and Pi engine adapters sharing a common agentLifecycle turn-management helper. Supporting CI/test infrastructure (under .github/drivers and .github/fixtures) wires these engines into a "three judges" evaluation fixture and CLI launch tooling, while integration tests in scripts/ exercise the real launcher and Copilot SDK connection when credentials are available.
Individual File Summaries
| File | Summary |
|---|---|
.agents/skills/rig/engines/anthropic.ts |
Implements the anthropicEngine AgentFactory, wrapping the Anthropic SDK beta JSON-schema tool-calling API with the shared agentLifecycle helper for turn management and tool execution. |
.agents/skills/rig/engines/codex.ts |
Implements the codexEngine AgentFactory wrapping the OpenAI Codex SDK, creating threads and delegating response-schema normalization to the core rig runtime. |
.agents/skills/rig/engines/deepseek.ts |
Implements the deepseekEngine AgentFactory using the DeepSeek Harness SDK client, reusing agentLifecycle for single-turn sequencing. |
.agents/skills/rig/engines/gemini.ts |
Implements the geminiEngine AgentFactory by spawning and communicating with a local Gemini CLI process rather than an SDK client. |
.agents/skills/rig/engines/pi.ts |
Implements the piEngine AgentFactory built on the Pi agent-core and pi-ai libraries, converting rig tool schemas into Pi tool definitions. |
.agents/skills/rig/engines/utils.ts |
Shared engine utility exporting agentLifecycle, which serializes concurrent turns per engine instance and tracks in-flight promises via an AbortController. |
.agents/skills/rig/globals.ts |
Provides ambient workflow helpers (call, pipeline, parallel) importable from "rig/globals" that delegate to the currently active workflow context for simpler porting of dynamic workflows. |
.agents/skills/rig/launch-tool.ts |
Defines createRigLaunchTool, a Copilot SDK Tool that spawns the rig launcher as a subprocess over a harness-owned connection to run a rig program source. |
.agents/skills/rig/rig.ts |
The core rig runtime (~3500 lines): declarative schema helpers, prompt intents, agent/workflow construction, sub-agent delegation, addons, and the Copilot SDK-backed engine plus CLI launcher. |
.agents/skills/rig/run.ts |
Tiny entrypoint script that imports and invokes runLauncherCli from rig.ts when executed directly as the installed skill binary. |
.github/drivers/codex-fixture-mcp.ts |
Fixture launcher that spawns a process and validates its three-judges-style assertion fixture for Codex-driver MCP integration testing. |
.github/drivers/copilot-sdk-driver.ts |
Driver script reading environment configuration and multi-provider JSON, then wiring the Copilot SDK with a harness-provided rig launch tool for CI integration. |
.github/drivers/pi-rig-extension.ts |
Defines a Pi agent-core tool extension that runs the rig CLI as a child process and validates results against the three-judges fixture. |
.github/fixtures/assert-three-judges.ts |
Shared test helper asserting a judging result object matches expected engine/model fields across engine integration fixtures. |
.github/fixtures/pi-gateway.ts |
Loads and validates the harness-provisioned Pi "aw-gateway" provider configuration and constructs Pi models/providers from it. |
.github/fixtures/provider-fixture.ts |
Builds a provider-engine fixture selecting among codex, deepseek, gemini, and pi engines based on required environment variables for three-judge testing. |
.github/fixtures/provider-three-judges.ts |
Thin entrypoint re-exporting createProviderFixture() as the default export for the provider three-judges integration test. |
.github/fixtures/three-judges.ts |
Defines the three-judges evaluation workflow: a judge agent configured per engine/model that scores a dummy request against clarity, safety, and feasibility criteria. |
scripts/haiku.integration.test.ts |
Vitest integration test that runs real haiku/sonnet/complex rig samples through the launcher CLI, skipped unless a Copilot token or SDK URI is configured. |
scripts/rig-launch-tool.integration.test.ts |
Live-only vitest test (gated by RIG_TOOL_INTEGRATION) that exercises createRigLaunchTool against a real CopilotClient/RuntimeConnection without exposing credential env vars. |
Note: This summary was generated by directly reading and analyzing the files listed above rather than by executing the rig program, because the sandboxed Copilot SDK connection token used by the harness is scoped to the parent process and is not available to a nested rig.ts invocation (AUTHENTICATION_FAILED). The project and per-file summaries above reflect the same 20 newest-sorted TypeScript source files (excluding node_modules and .git) that the find command in the rig program would have discovered. The sample input schema shorthand in the given program (input: { path: s.path }) also required the s.object({...}) wrapper to pass rig's schema validation.
Generated by Daily Rig File Summarizer · copilot · auto · 115.9 AIC · ⌖ 5.35 AIC · ⊞ 7.5K · ◷
- expires on Oct 18, 2026, 2:42 PM UTC
- Lenguaje dominante
- TypeScript
- Estrellas
- 13
- Forks
- 0
- Merge medio
- 3 h 44 min
- PR fusionados (30 d)
- 28
Preparar el entorno
Este proyecto no incluye contenedor de desarrollo, Dockerfile ni guía de contribución, así que la configuración corre por tu cuenta: empieza por su README y consulta nuestra guía para la primera contribución para los pasos generales.
Primeros pasos
- Lee el issue completo y luego la guía de contribución del proyecto.
- Comenta en el issue que vas a ocuparte — evita que dos personas hagan lo mismo.
- Haz un fork del repositorio y trabaja en una rama.
- Abre un pull request que haga referencia al número del issue.
Más de githubnext/rig
-
[aw] Upgrade availableAbiertoagentic-workflows
Dificultad 1/5 Menos de una hora Aptitud para principiantes 48/100
githubnext/rig#634 ·
Los mantenedores suelen responder en 1 día
-
agentic-workflows
Dificultad 4/5 3-5 días Aptitud para principiantes 35/100
githubnext/rig#633 · 1 comentario ·
Los mantenedores suelen responder en 1 día
-
agentic-workflows
Dificultad 4/5 3-5 días Aptitud para principiantes 30/100
githubnext/rig#631 ·
Los mantenedores suelen responder en 1 día
-
ai-agent automation
Dificultad 5/5 Más de una semana Aptitud para principiantes 1/100
githubnext/rig#630 ·
Los mantenedores suelen responder en 1 día
-
agentic-workflows
Dificultad 3/5 1-2 días Aptitud para principiantes 45/100
githubnext/rig#629 ·
Los mantenedores suelen responder en 1 día
Todos los issues de githubnext/rig
Issues similares
-
ui(ListToolbar): search takes the full row on a phone instead of sharing it with the result countAbierto
Dificultad 2/5 1-3 horas Aptitud para principiantes 64/100
Los mantenedores suelen responder en 1 día
-
priority:high
Dificultad 2/5 1-3 horas Aptitud para principiantes 72/100
Los mantenedores suelen responder en 1 día
-
Dificultad 2/5 1-3 horas Aptitud para principiantes 78/100
Los mantenedores suelen responder en 1 día
-
Dificultad 2/5 1-3 horas Aptitud para principiantes 65/100
immich-app/immich#32426 ·
Los mantenedores suelen responder en 1 día
-
documentation good first issue hacktoberfest help wanted
Dificultad 2/5 1-3 horas Aptitud para principiantes 78/100
libredb/libredb-studio#1756 ·
Los mantenedores suelen responder en 1 día