{"owner":"stanfordnlp","repo":"dspy","hasSkills":true,"totalSkillsCount":4,"totalTokensCount":819,"categories":["anthropic-skill"],"hasMcp":false,"mcpConfig":null,"found":["docs/docs/api/tools/PythonInterpreter.md","docs/docs/api/tools/index.md","docs/docs/api/tools/ColBERTv2.md","docs/docs/api/tools/Embeddings.md"],"skills":{"docs/docs/api/tools/PythonInterpreter.md":"# dspy.PythonInterpreter\n\n## Deno Installation\n\n`PythonInterpreter` uses Deno and Pyodide to run Python in a local WASM sandbox. The recommended installation\nkeeps Deno in the same Python environment as DSPy:\n\n```bash\npip install \"dspy[deno]\"\n```\n\nThe `deno` extra installs the official Deno Python distribution (`>=2.4.5,<3.0.0`). DSPy prefers that managed\nbinary when it is installed, so Python dependency locking also locks the Deno runtime. The extra provides binaries\nfor macOS x86-64/arm64, glibc Linux x86-64/arm64, and Windows x86-64 and adds approximately 40–50 MiB to the\nenvironment.\n\nOn other platforms, install a compatible Deno 2.x release (`>=2.0.0,<3.0.0`) using the\n[Deno installation instructions](https://docs.deno.com/runtime/getting_started/installation/). DSPy falls back to\nthe `deno` executable on `PATH`. An explicit `deno_command` passed to `PythonInterpreter` takes precedence over\nboth options.\n\nDSPy disables ambient `deno.json`, lockfile, `package.json`, and local `node_modules` discovery for its default\nrunner. A `package.json` in the current directory or an ancestor therefore cannot redirect the sandbox's pinned\nPyodide dependency. No `DENO_NO_PACKAGE_JSON` environment variable is required.\n\n## Execution Instructions\n\n`PythonInterpreter.execution_instructions` describes stable constraints of its Pyodide execution environment. It\nis class metadata, so code-generating modules can inspect it without starting Deno or allocating an interpreter.\n`RLM` includes these instructions in its action prompt, which adapters render in the model's system prompt.\n\nCustom interpreter factories may expose their own `execution_instructions` string. This metadata is optional; a\nfactory without it remains valid and uses RLM's generic action prompt.\n\n<!-- START_API_REF -->\n::: dspy.PythonInterpreter\n    handler: python\n    options:\n        members:\n            - __call__\n            - execute\n            - shutdown\n            - start\n        show_source: true\n        show_root_heading: true\n        heading_level: 2\n        docstring_style: google\n        show_root_full_path: true\n        show_object_full_path: false\n        separate_signature: false\n        inherited_members: true\n<!-- END_API_REF -->\n","docs/docs/api/tools/index.md":"# Tools\n\nAPI reference for DSPy tools. Select a page below.\n\n<!-- START_API_INDEX -->\n- [ColBERTv2](ColBERTv2.md)\n- [Embeddings](Embeddings.md)\n- [LocalInterpreter](LocalInterpreter.md)\n- [PythonInterpreter](PythonInterpreter.md)\n<!-- END_API_INDEX -->\n","docs/docs/api/tools/ColBERTv2.md":"# dspy.ColBERTv2\n\n<!-- START_API_REF -->\n::: dspy.ColBERTv2\n    handler: python\n    options:\n        members:\n            - __call__\n        show_source: true\n        show_root_heading: true\n        heading_level: 2\n        docstring_style: google\n        show_root_full_path: true\n        show_object_full_path: false\n        separate_signature: false\n        inherited_members: true\n<!-- END_API_REF -->\n","docs/docs/api/tools/Embeddings.md":"# dspy.retrievers.Embeddings\n\n!!! note \"Requires numpy\"\n    `dspy.retrievers.Embeddings` requires numpy. Install it with `pip install dspy[numpy]`.\n\n<!-- START_API_REF -->\n::: dspy.Embeddings\n    handler: python\n    options:\n        members:\n            - __call__\n            - forward\n            - from_saved\n            - load\n            - save\n        show_source: true\n        show_root_heading: true\n        heading_level: 2\n        docstring_style: google\n        show_root_full_path: true\n        show_object_full_path: false\n        separate_signature: false\n        inherited_members: true\n<!-- END_API_REF -->\n"},"files":{"docs/docs/api/tools/PythonInterpreter.md":"# dspy.PythonInterpreter\n\n## Deno Installation\n\n`PythonInterpreter` uses Deno and Pyodide to run Python in a local WASM sandbox. The recommended installation\nkeeps Deno in the same Python environment as DSPy:\n\n```bash\npip install \"dspy[deno]\"\n```\n\nThe `deno` extra installs the official Deno Python distribution (`>=2.4.5,<3.0.0`). DSPy prefers that managed\nbinary when it is installed, so Python dependency locking also locks the Deno runtime. The extra provides binaries\nfor macOS x86-64/arm64, glibc Linux x86-64/arm64, and Windows x86-64 and adds approximately 40–50 MiB to the\nenvironment.\n\nOn other platforms, install a compatible Deno 2.x release (`>=2.0.0,<3.0.0`) using the\n[Deno installation instructions](https://docs.deno.com/runtime/getting_started/installation/). DSPy falls back to\nthe `deno` executable on `PATH`. An explicit `deno_command` passed to `PythonInterpreter` takes precedence over\nboth options.\n\nDSPy disables ambient `deno.json`, lockfile, `package.json`, and local `node_modules` discovery for its default\nrunner. A `package.json` in the current directory or an ancestor therefore cannot redirect the sandbox's pinned\nPyodide dependency. No `DENO_NO_PACKAGE_JSON` environment variable is required.\n\n## Execution Instructions\n\n`PythonInterpreter.execution_instructions` describes stable constraints of its Pyodide execution environment. It\nis class metadata, so code-generating modules can inspect it without starting Deno or allocating an interpreter.\n`RLM` includes these instructions in its action prompt, which adapters render in the model's system prompt.\n\nCustom interpreter factories may expose their own `execution_instructions` string. This metadata is optional; a\nfactory without it remains valid and uses RLM's generic action prompt.\n\n<!-- START_API_REF -->\n::: dspy.PythonInterpreter\n    handler: python\n    options:\n        members:\n            - __call__\n            - execute\n            - shutdown\n            - start\n        show_source: true\n        show_root_heading: true\n        heading_level: 2\n        docstring_style: google\n        show_root_full_path: true\n        show_object_full_path: false\n        separate_signature: false\n        inherited_members: true\n<!-- END_API_REF -->\n","docs/docs/api/tools/index.md":"# Tools\n\nAPI reference for DSPy tools. Select a page below.\n\n<!-- START_API_INDEX -->\n- [ColBERTv2](ColBERTv2.md)\n- [Embeddings](Embeddings.md)\n- [LocalInterpreter](LocalInterpreter.md)\n- [PythonInterpreter](PythonInterpreter.md)\n<!-- END_API_INDEX -->\n","docs/docs/api/tools/ColBERTv2.md":"# dspy.ColBERTv2\n\n<!-- START_API_REF -->\n::: dspy.ColBERTv2\n    handler: python\n    options:\n        members:\n            - __call__\n        show_source: true\n        show_root_heading: true\n        heading_level: 2\n        docstring_style: google\n        show_root_full_path: true\n        show_object_full_path: false\n        separate_signature: false\n        inherited_members: true\n<!-- END_API_REF -->\n","docs/docs/api/tools/Embeddings.md":"# dspy.retrievers.Embeddings\n\n!!! note \"Requires numpy\"\n    `dspy.retrievers.Embeddings` requires numpy. Install it with `pip install dspy[numpy]`.\n\n<!-- START_API_REF -->\n::: dspy.Embeddings\n    handler: python\n    options:\n        members:\n            - __call__\n            - forward\n            - from_saved\n            - load\n            - save\n        show_source: true\n        show_root_heading: true\n        heading_level: 2\n        docstring_style: google\n        show_root_full_path: true\n        show_object_full_path: false\n        separate_signature: false\n        inherited_members: true\n<!-- END_API_REF -->\n"},"items":[{"name":"ColBERTv2.md","path":"docs/docs/api/tools/ColBERTv2.md","rawUrl":"https://raw.githubusercontent.com/stanfordnlp/dspy/HEAD/docs/docs/api/tools/ColBERTv2.md","title":"Tools Skill","category":"anthropic-skill","format":"markdown","content":"# dspy.ColBERTv2\n\n<!-- START_API_REF -->\n::: dspy.ColBERTv2\n    handler: python\n    options:\n        members:\n            - __call__\n        show_source: true\n        show_root_heading: true\n        heading_level: 2\n        docstring_style: google\n        show_root_full_path: true\n        show_object_full_path: false\n        separate_signature: false\n        inherited_members: true\n<!-- END_API_REF -->\n","isInternal":false,"tokens":98,"sizeBytes":406},{"name":"Embeddings.md","path":"docs/docs/api/tools/Embeddings.md","rawUrl":"https://raw.githubusercontent.com/stanfordnlp/dspy/HEAD/docs/docs/api/tools/Embeddings.md","title":"Tools Skill","category":"anthropic-skill","format":"markdown","content":"# dspy.retrievers.Embeddings\n\n!!! note \"Requires numpy\"\n    `dspy.retrievers.Embeddings` requires numpy. Install it with `pip install dspy[numpy]`.\n\n<!-- START_API_REF -->\n::: dspy.Embeddings\n    handler: python\n    options:\n        members:\n            - __call__\n            - forward\n            - from_saved\n            - load\n            - save\n        show_source: true\n        show_root_heading: true\n        heading_level: 2\n        docstring_style: google\n        show_root_full_path: true\n        show_object_full_path: false\n        separate_signature: false\n        inherited_members: true\n<!-- END_API_REF -->\n","isInternal":false,"tokens":145,"sizeBytes":623},{"name":"index.md","path":"docs/docs/api/tools/index.md","rawUrl":"https://raw.githubusercontent.com/stanfordnlp/dspy/HEAD/docs/docs/api/tools/index.md","title":"Tools Skill","category":"anthropic-skill","format":"markdown","content":"# Tools\n\nAPI reference for DSPy tools. Select a page below.\n\n<!-- START_API_INDEX -->\n- [ColBERTv2](ColBERTv2.md)\n- [Embeddings](Embeddings.md)\n- [LocalInterpreter](LocalInterpreter.md)\n- [PythonInterpreter](PythonInterpreter.md)\n<!-- END_API_INDEX -->\n","isInternal":false,"tokens":65,"sizeBytes":253},{"name":"PythonInterpreter.md","path":"docs/docs/api/tools/PythonInterpreter.md","rawUrl":"https://raw.githubusercontent.com/stanfordnlp/dspy/HEAD/docs/docs/api/tools/PythonInterpreter.md","title":"Tools Skill","category":"anthropic-skill","format":"markdown","content":"# dspy.PythonInterpreter\n\n## Deno Installation\n\n`PythonInterpreter` uses Deno and Pyodide to run Python in a local WASM sandbox. The recommended installation\nkeeps Deno in the same Python environment as DSPy:\n\n```bash\npip install \"dspy[deno]\"\n```\n\nThe `deno` extra installs the official Deno Python distribution (`>=2.4.5,<3.0.0`). DSPy prefers that managed\nbinary when it is installed, so Python dependency locking also locks the Deno runtime. The extra provides binaries\nfor macOS x86-64/arm64, glibc Linux x86-64/arm64, and Windows x86-64 and adds approximately 40–50 MiB to the\nenvironment.\n\nOn other platforms, install a compatible Deno 2.x release (`>=2.0.0,<3.0.0`) using the\n[Deno installation instructions](https://docs.deno.com/runtime/getting_started/installation/). DSPy falls back to\nthe `deno` executable on `PATH`. An explicit `deno_command` passed to `PythonInterpreter` takes precedence over\nboth options.\n\nDSPy disables ambient `deno.json`, lockfile, `package.json`, and local `node_modules` discovery for its default\nrunner. A `package.json` in the current directory or an ancestor therefore cannot redirect the sandbox's pinned\nPyodide dependency. No `DENO_NO_PACKAGE_JSON` environment variable is required.\n\n## Execution Instructions\n\n`PythonInterpreter.execution_instructions` describes stable constraints of its Pyodide execution environment. It\nis class metadata, so code-generating modules can inspect it without starting Deno or allocating an interpreter.\n`RLM` includes these instructions in its action prompt, which adapters render in the model's system prompt.\n\nCustom interpreter factories may expose their own `execution_instructions` string. This metadata is optional; a\nfactory without it remains valid and uses RLM's generic action prompt.\n\n<!-- START_API_REF -->\n::: dspy.PythonInterpreter\n    handler: python\n    options:\n        members:\n            - __call__\n            - execute\n            - shutdown\n            - start\n        show_source: true\n        show_root_heading: true\n        heading_level: 2\n        docstring_style: google\n        show_root_full_path: true\n        show_object_full_path: false\n        separate_signature: false\n        inherited_members: true\n<!-- END_API_REF -->\n","isInternal":false,"tokens":511,"sizeBytes":2238}],"systemPromptSnippet":"<yakaai_skills repo=\"stanfordnlp/dspy\">\n<!-- File: docs/docs/api/tools/ColBERTv2.md (Tokens: ~98 | Category: anthropic-skill) -->\n# dspy.ColBERTv2\n\n<!-- START_API_REF -->\n::: dspy.ColBERTv2\n    handler: python\n    options:\n        members:\n            - __call__\n        show_source: true\n        show_root_heading: true\n        heading_level: 2\n        docstring_style: google\n        show_root_full_path: true\n        show_object_full_path: false\n        separate_signature: false\n        inherited_members: true\n<!-- END_API_REF -->\n\n<!-- File: docs/docs/api/tools/Embeddings.md (Tokens: ~145 | Category: anthropic-skill) -->\n# dspy.retrievers.Embeddings\n\n!!! note \"Requires numpy\"\n    `dspy.retrievers.Embeddings` requires numpy. Install it with `pip install dspy[numpy]`.\n\n<!-- START_API_REF -->\n::: dspy.Embeddings\n    handler: python\n    options:\n        members:\n            - __call__\n            - forward\n            - from_saved\n            - load\n            - save\n        show_source: true\n        show_root_heading: true\n        heading_level: 2\n        docstring_style: google\n        show_root_full_path: true\n        show_object_full_path: false\n        separate_signature: false\n        inherited_members: true\n<!-- END_API_REF -->\n\n<!-- File: docs/docs/api/tools/index.md (Tokens: ~65 | Category: anthropic-skill) -->\n# Tools\n\nAPI reference for DSPy tools. Select a page below.\n\n<!-- START_API_INDEX -->\n- [ColBERTv2](ColBERTv2.md)\n- [Embeddings](Embeddings.md)\n- [LocalInterpreter](LocalInterpreter.md)\n- [PythonInterpreter](PythonInterpreter.md)\n<!-- END_API_INDEX -->\n\n<!-- File: docs/docs/api/tools/PythonInterpreter.md (Tokens: ~511 | Category: anthropic-skill) -->\n# dspy.PythonInterpreter\n\n## Deno Installation\n\n`PythonInterpreter` uses Deno and Pyodide to run Python in a local WASM sandbox. The recommended installation\nkeeps Deno in the same Python environment as DSPy:\n\n```bash\npip install \"dspy[deno]\"\n```\n\nThe `deno` extra installs the official Deno Python distribution (`>=2.4.5,<3.0.0`). DSPy prefers that managed\nbinary when it is installed, so Python dependency locking also locks the Deno runtime. The extra provides binaries\nfor macOS x86-64/arm64, glibc Linux x86-64/arm64, and Windows x86-64 and adds approximately 40–50 MiB to the\nenvironment.\n\nOn other platforms, install a compatible Deno 2.x release (`>=2.0.0,<3.0.0`) using the\n[Deno installation instructions](https://docs.deno.com/runtime/getting_started/installation/). DSPy falls back to\nthe `deno` executable on `PATH`. An explicit `deno_command` passed to `PythonInterpreter` takes precedence over\nboth options.\n\nDSPy disables ambient `deno.json`, lockfile, `package.json`, and local `node_modules` discovery for its default\nrunner. A `package.json` in the current directory or an ancestor therefore cannot redirect the sandbox's pinned\nPyodide dependency. No `DENO_NO_PACKAGE_JSON` environment variable is required.\n\n## Execution Instructions\n\n`PythonInterpreter.execution_instructions` describes stable constraints of its Pyodide execution environment. It\nis class metadata, so code-generating modules can inspect it without starting Deno or allocating an interpreter.\n`RLM` includes these instructions in its action prompt, which adapters render in the model's system prompt.\n\nCustom interpreter factories may expose their own `execution_instructions` string. This metadata is optional; a\nfactory without it remains valid and uses RLM's generic action prompt.\n\n<!-- START_API_REF -->\n::: dspy.PythonInterpreter\n    handler: python\n    options:\n        members:\n            - __call__\n            - execute\n            - shutdown\n            - start\n        show_source: true\n        show_root_heading: true\n        heading_level: 2\n        docstring_style: google\n        show_root_full_path: true\n        show_object_full_path: false\n        separate_signature: false\n        inherited_members: true\n<!-- END_API_REF -->\n</yakaai_skills>"}