PyPI - agentledger-runtime - Versions diffs - 1.2.2__tar.gz → 1.2.4__tar.gz - Mend

agentledger-runtime 1.2.2tar.gz → 1.2.4tar.gz

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.

Files changed (369) hide show

{agentledger_runtime-1.2.2 → agentledger_runtime-1.2.4}/.gitignore RENAMED Viewed

@@ -1,4 +1,6 @@
 .agentledger/
+go/.agentledger-go/
+rust/.agentledger-rust/
 *.db
 *.sqlite
 *.sqlite3

{agentledger_runtime-1.2.2 → agentledger_runtime-1.2.4}/CHANGELOG.md RENAMED Viewed

@@ -8,6 +8,43 @@ This project follows semantic versioning for the stable runtime-core contract. O
 No unreleased changes yet.
+## 1.2.4
+### Added
+- Open-source impact documentation for AgentLedger's early-stage agent reliability and governance infrastructure positioning.
+- Maintainer notes documenting project ownership, review principles, release responsibilities, coding-agent usage, and maintenance signals.
+- Cross-language 3-minute side-effect safety demos showing crash/retry without duplicate external writes.
+- Cross-language MCP governance examples showing descriptor annotations for side effects, approval, sandbox metadata, idempotency, and audit evidence.
+- Adoption plan, public issue/discussion candidates, and a legal-agent case study template.
+### Changed
+- README first screen now leads with the tool side-effect safety problem before describing architecture.
+- MCP tool adapter now maps governance annotations such as `approval_required`, `sandbox_required`, `sandbox_executor`, and `sandbox_policy`.
+### Clarified
+- Roadmap guidance for open-source adoption, OpenAI Agents SDK/MCP examples, Codex-assisted maintainer workflows, and public usage evidence without changing runtime-core semantics.
+- 1.2.4 does not change runtime-core semantics; it is an adoption and example-focused release.
+## 1.2.3
+### Added
+- SQL query examples for runtime inspection, multi-table timelines, approvals, costs, artifacts, and large business-schema integration patterns.
+- `agentledger-langfuse` as an official optional observability adapter boundary.
+- TypeScript subpath/package, Go adapter boundary, and Rust crate/feature boundary for Langfuse-style evidence/trace payload export.
+### Changed
+- Updated adapter packaging, certification, optional-adapter conformance, and documentation entrypoints for the Langfuse adapter boundary.
+- Removed tracked local runtime state from the repository.
+### Clarified
+- Langfuse support is an adapter/export boundary, not a replacement for Langfuse or a binding to the Langfuse SDK.
 ## 1.2.2
 ### Added

{agentledger_runtime-1.2.2 → agentledger_runtime-1.2.4}/PKG-INFO RENAMED Viewed

@@ -1,7 +1,7 @@
 Metadata-Version: 2.4
 Name: agentledger-runtime
-Version: 1.2.2
-Summary: Durable execution and reliability runtime for production-grade AI agents
+Version: 1.2.4
+Summary: Runtime reliability layer for Agent Harness stacks
 Project-URL: Homepage, https://github.com/yaogdu/AgentLedger
 Project-URL: Repository, https://github.com/yaogdu/AgentLedger
 Project-URL: Documentation, https://github.com/yaogdu/AgentLedger#readme
@@ -21,6 +21,7 @@ Classifier: Programming Language :: Python :: 3.12
 Classifier: Topic :: Software Development :: Libraries :: Python Modules
 Requires-Python: >=3.11
 Provides-Extra: all
+Requires-Dist: agentledger-langfuse<2,>=1.2; extra == 'all'
 Requires-Dist: agentledger-langgraph<2,>=1.2; extra == 'all'
 Requires-Dist: agentledger-mcp<2,>=1.2; extra == 'all'
 Requires-Dist: agentledger-mysql<2,>=1.2; extra == 'all'
@@ -30,6 +31,8 @@ Requires-Dist: agentledger-s3<2,>=1.2; extra == 'all'
 Requires-Dist: agentledger-sandbox-docker<2,>=1.2; extra == 'all'
 Provides-Extra: docker
 Requires-Dist: agentledger-sandbox-docker<2,>=1.2; extra == 'docker'
+Provides-Extra: langfuse
+Requires-Dist: agentledger-langfuse<2,>=1.2; extra == 'langfuse'
 Provides-Extra: langgraph
 Requires-Dist: agentledger-langgraph<2,>=1.2; extra == 'langgraph'
 Provides-Extra: mcp
@@ -49,17 +52,21 @@ Description-Content-Type: text/markdown
 [English](README.md) | [中文](README.zh-CN.md)
 ![Python 3.11+](https://img.shields.io/badge/Python-3.11%2B-3776AB?logo=python&logoColor=white)
-![Version 1.2.2 stable](https://img.shields.io/badge/Version-1.2.2--stable-111827)
+![Version 1.2.4 stable](https://img.shields.io/badge/Version-1.2.4--stable-111827)
 ![License Apache-2.0](https://img.shields.io/badge/License-Apache--2.0-0f766e)
 ![Runtime Durable](https://img.shields.io/badge/Runtime-durable%20execution-1f6feb)
 ![Storage SQLite/Postgres/MySQL](https://img.shields.io/badge/Storage-SQLite%20%7C%20Postgres%20%7C%20MySQL-b45309)
 ![Replay Evidence](https://img.shields.io/badge/Replay-evidence%20driven-7c3aed)
 ![Tool Ledger](https://img.shields.io/badge/Tools-ledger%20guarded-d97706)
-AgentLedger `1.2.2` is an agent execution safety, evidence, and reliability layer. It does not try to teach agents how to reason; it makes agent runs durable, auditable, replayable, policy-governed, and recoverable when workers crash, tools fail, or prompts change.
+Your agent called a tool. Did it happen? Can you retry safely? Can you prove it later?
+AgentLedger `1.2.4` is a runtime reliability layer for Agent Harness stacks. It does not try to teach agents how to reason or replace the surrounding harness ecosystem; it makes agent runs durable, auditable, replayable, policy-governed, and recoverable when workers crash, tools fail, or prompts change.
 Most agent frameworks focus on planning, reasoning, and workflow logic. AgentLedger sits underneath or beside LangChain, LangGraph, CrewAI, AutoGen, OpenAI Agents SDK, LlamaIndex, Semantic Kernel, or custom agents to provide runtime guarantees around state, tools, evidence, replay, and recovery.
+In a full harness stack, systems such as LangGraph, Temporal, Langfuse, MCP, model routers, storage backends, and sandbox providers can each own the layer they are good at. AgentLedger owns the reliability substrate between them: durable execution, tool/model governance, evidence, replay, policy/sandbox boundaries, cost/failure attribution, and adapter contracts.
 Python remains the reference implementation, and Go, TypeScript, and Rust now have native runtime-core baselines aligned to the same language-neutral runtime contract. Provider-specific drivers and framework-native adapters intentionally vary by ecosystem. See `docs/LANGUAGE_IMPLEMENTATION_COMPARISON.md` for the exact four-language comparison and adapter boundary.
 ## Start Here
@@ -69,6 +76,11 @@ Python remains the reference implementation, and Go, TypeScript, and Rust now ha
 | Install and run the first example | [docs/GETTING_STARTED.md](docs/GETTING_STARTED.md) |
 | Choose Python / Go / TypeScript / Rust | [docs/LANGUAGE_QUICKSTART.md](docs/LANGUAGE_QUICKSTART.md) |
 | Find runnable examples | [examples/README.md](examples/README.md), [go/examples/README.md](go/examples/README.md), [typescript/examples/README.md](typescript/examples/README.md), [rust/examples/README.md](rust/examples/README.md) |
+| Query runtime tables | [docs/QUERY_EXAMPLES.md](docs/QUERY_EXAMPLES.md) |
+| Understand Harness stack composition | [docs/HARNESS_STACK.md](docs/HARNESS_STACK.md) |
+| Understand open-source impact | [docs/OPEN_SOURCE_IMPACT.md](docs/OPEN_SOURCE_IMPACT.md) |
+| Understand maintainer responsibilities | [docs/MAINTAINER_NOTES.md](docs/MAINTAINER_NOTES.md) |
+| Plan adoption work | [docs/ADOPTION.md](docs/ADOPTION.md) |
 | Understand what is equal across languages | [docs/LANGUAGE_IMPLEMENTATION_COMPARISON.md](docs/LANGUAGE_IMPLEMENTATION_COMPARISON.md) |
 | Install optional adapter packages | [docs/ADAPTER_PACKAGING.md](docs/ADAPTER_PACKAGING.md) |
 | Use Go correctly | [go/README.md](go/README.md#install) |
@@ -117,6 +129,23 @@ Optional production adapter:
 For example, sandbox semantics are core, but sandbox infrastructure is not. Core owns `SandboxPolicy`, fail-closed routing, audit/evidence records, and replay safety; Docker, E2B, bubblewrap, Kubernetes/gVisor, Firecracker, or custom executors are adapters.
+## Harness stack composition
+AgentLedger is not a full Agent Harness by itself. It is designed to compose with the rest of the harness ecosystem:
+For concrete stack patterns, from a minimal local harness to a Temporal + LangGraph + observability harness, see [docs/HARNESS_STACK.md](docs/HARNESS_STACK.md).
+| Harness layer | Typical systems | AgentLedger role |
+| --- | --- | --- |
+| Agent workflow / planning | LangGraph, LangChain, CrewAI, AutoGen, OpenAI Agents SDK, custom code | wrap nodes and tools with durable state, policy, Tool Ledger, evidence, and replay guarantees |
+| Durable orchestration | Temporal, Ray, Kubernetes workers | provide agent-specific leases, fencing, checkpoints, cancellation, cost/failure attribution, and replay semantics inside the worker step |
+| Observability / eval UI | Langfuse, LangSmith, OpenTelemetry, custom dashboards | export structured runtime events, evidence bundles, trace/cost/failure data, and correlation IDs |
+| Tool and context protocols | MCP, internal tool servers, provider SDK tools | enforce schema, permissions, approval, sandbox, idempotency, and audit before side effects happen |
+| Model providers / routers | OpenAI, Anthropic, Gemini, Bedrock, Ollama, LiteLLM, enterprise gateways | provide the runtime model-call contract, archived responses, budget/fallback/replay semantics, and optional provider adapters |
+| Storage / artifacts | SQLite, Postgres, MySQL, S3/MinIO, internal stores | keep runtime metadata, state versions, migrations, blob refs, and evidence refs durable and conformance-tested |
+The intended production shape is therefore not `AgentLedger instead of LangGraph/Temporal/Langfuse`. It is `AgentLedger with LangGraph/Temporal/Langfuse` where AgentLedger governs the model/tool/state boundary that those systems otherwise cannot enforce by themselves.
 ## What AgentLedger is for
 - Making long-running agent tasks resume from the last committed checkpoint after crash or restart
@@ -140,10 +169,12 @@ For example, sandbox semantics are core, but sandbox infrastructure is not. Core
 ## Examples
-The repository includes small quickstarts plus a richer multi-language Travel Assistant demo that shows the same runtime ideas across Python, Go, Rust, and TypeScript.
+The repository includes cross-language 3-minute side-effect safety demos, MCP governance demos, small quickstarts, and a richer multi-language Travel Assistant demo that shows the same runtime ideas across Python, Go, Rust, and TypeScript.
-| Language | Demo | Run |
+| Goal | Demo | Run |
 | --- | --- | --- |
+| 3-minute side-effect safety | Python / Go / Rust / TypeScript | `PYTHONPATH=src python3 examples/three_minute_demo/demo.py`; `cd go && go run ./examples/three_minute_demo`; `cd rust && cargo run --example three_minute_demo`; `cd typescript && node examples/three_minute_demo/three_minute_demo.js` |
+| MCP tool governance | Python / Go / Rust / TypeScript | `PYTHONPATH=src python3 examples/mcp_governance/demo.py`; `cd go && go run ./examples/mcp_governance`; `cd rust && cargo run --example mcp_governance`; `cd typescript && node examples/mcp_governance/mcp_governance.js` |
 | Python | `examples/travel_assistant/demo.py` | `python3 examples/travel_assistant/demo.py` |
 | Go | `go/examples/travel_assistant/main.go` | `cd go && go run examples/travel_assistant/main.go` |
 | Rust | `rust/examples/travel_assistant.rs` | `cd rust && cargo run --example travel_assistant` |
@@ -298,7 +329,7 @@ AgentLedger is also not a new LLM SDK, not a workflow engine, not a general obse
 ## Current maturity
-AgentLedger 1.2.2 is a stable runtime-core release with Python as the reference implementation and Go, TypeScript, and Rust covered by shared runtime-core parity gates. It is suitable for local use, framework adapter integration, reliability semantics validation, and production pilot preparation with explicit adapter boundaries.
+AgentLedger 1.2.4 is a stable runtime-core release with Python as the reference implementation and Go, TypeScript, and Rust covered by shared runtime-core parity gates. It is suitable for local use, framework adapter integration, reliability semantics validation, and production pilot preparation with explicit adapter boundaries.
 The runtime-core contract is stable; optional production adapters and external infrastructure hardening remain separately tracked. See [docs/MATURITY_MODEL.md](docs/MATURITY_MODEL.md), [docs/IMPLEMENTATION_STATUS.md](docs/IMPLEMENTATION_STATUS.md), and [docs/ROADMAP.md](docs/ROADMAP.md).

{agentledger_runtime-1.2.2 → agentledger_runtime-1.2.4}/README.md RENAMED Viewed

@@ -3,17 +3,21 @@
 [English](README.md) | [中文](README.zh-CN.md)
 ![Python 3.11+](https://img.shields.io/badge/Python-3.11%2B-3776AB?logo=python&logoColor=white)
-![Version 1.2.2 stable](https://img.shields.io/badge/Version-1.2.2--stable-111827)
+![Version 1.2.4 stable](https://img.shields.io/badge/Version-1.2.4--stable-111827)
 ![License Apache-2.0](https://img.shields.io/badge/License-Apache--2.0-0f766e)
 ![Runtime Durable](https://img.shields.io/badge/Runtime-durable%20execution-1f6feb)
 ![Storage SQLite/Postgres/MySQL](https://img.shields.io/badge/Storage-SQLite%20%7C%20Postgres%20%7C%20MySQL-b45309)
 ![Replay Evidence](https://img.shields.io/badge/Replay-evidence%20driven-7c3aed)
 ![Tool Ledger](https://img.shields.io/badge/Tools-ledger%20guarded-d97706)
-AgentLedger `1.2.2` is an agent execution safety, evidence, and reliability layer. It does not try to teach agents how to reason; it makes agent runs durable, auditable, replayable, policy-governed, and recoverable when workers crash, tools fail, or prompts change.
+Your agent called a tool. Did it happen? Can you retry safely? Can you prove it later?
+AgentLedger `1.2.4` is a runtime reliability layer for Agent Harness stacks. It does not try to teach agents how to reason or replace the surrounding harness ecosystem; it makes agent runs durable, auditable, replayable, policy-governed, and recoverable when workers crash, tools fail, or prompts change.
 Most agent frameworks focus on planning, reasoning, and workflow logic. AgentLedger sits underneath or beside LangChain, LangGraph, CrewAI, AutoGen, OpenAI Agents SDK, LlamaIndex, Semantic Kernel, or custom agents to provide runtime guarantees around state, tools, evidence, replay, and recovery.
+In a full harness stack, systems such as LangGraph, Temporal, Langfuse, MCP, model routers, storage backends, and sandbox providers can each own the layer they are good at. AgentLedger owns the reliability substrate between them: durable execution, tool/model governance, evidence, replay, policy/sandbox boundaries, cost/failure attribution, and adapter contracts.
 Python remains the reference implementation, and Go, TypeScript, and Rust now have native runtime-core baselines aligned to the same language-neutral runtime contract. Provider-specific drivers and framework-native adapters intentionally vary by ecosystem. See `docs/LANGUAGE_IMPLEMENTATION_COMPARISON.md` for the exact four-language comparison and adapter boundary.
 ## Start Here
@@ -23,6 +27,11 @@ Python remains the reference implementation, and Go, TypeScript, and Rust now ha
 | Install and run the first example | [docs/GETTING_STARTED.md](docs/GETTING_STARTED.md) |
 | Choose Python / Go / TypeScript / Rust | [docs/LANGUAGE_QUICKSTART.md](docs/LANGUAGE_QUICKSTART.md) |
 | Find runnable examples | [examples/README.md](examples/README.md), [go/examples/README.md](go/examples/README.md), [typescript/examples/README.md](typescript/examples/README.md), [rust/examples/README.md](rust/examples/README.md) |
+| Query runtime tables | [docs/QUERY_EXAMPLES.md](docs/QUERY_EXAMPLES.md) |
+| Understand Harness stack composition | [docs/HARNESS_STACK.md](docs/HARNESS_STACK.md) |
+| Understand open-source impact | [docs/OPEN_SOURCE_IMPACT.md](docs/OPEN_SOURCE_IMPACT.md) |
+| Understand maintainer responsibilities | [docs/MAINTAINER_NOTES.md](docs/MAINTAINER_NOTES.md) |
+| Plan adoption work | [docs/ADOPTION.md](docs/ADOPTION.md) |
 | Understand what is equal across languages | [docs/LANGUAGE_IMPLEMENTATION_COMPARISON.md](docs/LANGUAGE_IMPLEMENTATION_COMPARISON.md) |
 | Install optional adapter packages | [docs/ADAPTER_PACKAGING.md](docs/ADAPTER_PACKAGING.md) |
 | Use Go correctly | [go/README.md](go/README.md#install) |
@@ -71,6 +80,23 @@ Optional production adapter:
 For example, sandbox semantics are core, but sandbox infrastructure is not. Core owns `SandboxPolicy`, fail-closed routing, audit/evidence records, and replay safety; Docker, E2B, bubblewrap, Kubernetes/gVisor, Firecracker, or custom executors are adapters.
+## Harness stack composition
+AgentLedger is not a full Agent Harness by itself. It is designed to compose with the rest of the harness ecosystem:
+For concrete stack patterns, from a minimal local harness to a Temporal + LangGraph + observability harness, see [docs/HARNESS_STACK.md](docs/HARNESS_STACK.md).
+| Harness layer | Typical systems | AgentLedger role |
+| --- | --- | --- |
+| Agent workflow / planning | LangGraph, LangChain, CrewAI, AutoGen, OpenAI Agents SDK, custom code | wrap nodes and tools with durable state, policy, Tool Ledger, evidence, and replay guarantees |
+| Durable orchestration | Temporal, Ray, Kubernetes workers | provide agent-specific leases, fencing, checkpoints, cancellation, cost/failure attribution, and replay semantics inside the worker step |
+| Observability / eval UI | Langfuse, LangSmith, OpenTelemetry, custom dashboards | export structured runtime events, evidence bundles, trace/cost/failure data, and correlation IDs |
+| Tool and context protocols | MCP, internal tool servers, provider SDK tools | enforce schema, permissions, approval, sandbox, idempotency, and audit before side effects happen |
+| Model providers / routers | OpenAI, Anthropic, Gemini, Bedrock, Ollama, LiteLLM, enterprise gateways | provide the runtime model-call contract, archived responses, budget/fallback/replay semantics, and optional provider adapters |
+| Storage / artifacts | SQLite, Postgres, MySQL, S3/MinIO, internal stores | keep runtime metadata, state versions, migrations, blob refs, and evidence refs durable and conformance-tested |
+The intended production shape is therefore not `AgentLedger instead of LangGraph/Temporal/Langfuse`. It is `AgentLedger with LangGraph/Temporal/Langfuse` where AgentLedger governs the model/tool/state boundary that those systems otherwise cannot enforce by themselves.
 ## What AgentLedger is for
 - Making long-running agent tasks resume from the last committed checkpoint after crash or restart
@@ -94,10 +120,12 @@ For example, sandbox semantics are core, but sandbox infrastructure is not. Core
 ## Examples
-The repository includes small quickstarts plus a richer multi-language Travel Assistant demo that shows the same runtime ideas across Python, Go, Rust, and TypeScript.
+The repository includes cross-language 3-minute side-effect safety demos, MCP governance demos, small quickstarts, and a richer multi-language Travel Assistant demo that shows the same runtime ideas across Python, Go, Rust, and TypeScript.
-| Language | Demo | Run |
+| Goal | Demo | Run |
 | --- | --- | --- |
+| 3-minute side-effect safety | Python / Go / Rust / TypeScript | `PYTHONPATH=src python3 examples/three_minute_demo/demo.py`; `cd go && go run ./examples/three_minute_demo`; `cd rust && cargo run --example three_minute_demo`; `cd typescript && node examples/three_minute_demo/three_minute_demo.js` |
+| MCP tool governance | Python / Go / Rust / TypeScript | `PYTHONPATH=src python3 examples/mcp_governance/demo.py`; `cd go && go run ./examples/mcp_governance`; `cd rust && cargo run --example mcp_governance`; `cd typescript && node examples/mcp_governance/mcp_governance.js` |
 | Python | `examples/travel_assistant/demo.py` | `python3 examples/travel_assistant/demo.py` |
 | Go | `go/examples/travel_assistant/main.go` | `cd go && go run examples/travel_assistant/main.go` |
 | Rust | `rust/examples/travel_assistant.rs` | `cd rust && cargo run --example travel_assistant` |
@@ -252,7 +280,7 @@ AgentLedger is also not a new LLM SDK, not a workflow engine, not a general obse
 ## Current maturity
-AgentLedger 1.2.2 is a stable runtime-core release with Python as the reference implementation and Go, TypeScript, and Rust covered by shared runtime-core parity gates. It is suitable for local use, framework adapter integration, reliability semantics validation, and production pilot preparation with explicit adapter boundaries.
+AgentLedger 1.2.4 is a stable runtime-core release with Python as the reference implementation and Go, TypeScript, and Rust covered by shared runtime-core parity gates. It is suitable for local use, framework adapter integration, reliability semantics validation, and production pilot preparation with explicit adapter boundaries.
 The runtime-core contract is stable; optional production adapters and external infrastructure hardening remain separately tracked. See [docs/MATURITY_MODEL.md](docs/MATURITY_MODEL.md), [docs/IMPLEMENTATION_STATUS.md](docs/IMPLEMENTATION_STATUS.md), and [docs/ROADMAP.md](docs/ROADMAP.md).

{agentledger_runtime-1.2.2 → agentledger_runtime-1.2.4}/README.zh-CN.md RENAMED Viewed

@@ -3,17 +3,21 @@
 [English](README.md) | [中文](README.zh-CN.md)
 ![Python 3.11+](https://img.shields.io/badge/Python-3.11%2B-3776AB?logo=python&logoColor=white)
-![Version 1.2.2 stable](https://img.shields.io/badge/Version-1.2.2--stable-111827)
+![Version 1.2.4 stable](https://img.shields.io/badge/Version-1.2.4--stable-111827)
 ![License Apache-2.0](https://img.shields.io/badge/License-Apache--2.0-0f766e)
 ![Runtime Durable](https://img.shields.io/badge/Runtime-durable%20execution-1f6feb)
 ![Storage SQLite/Postgres/MySQL](https://img.shields.io/badge/Storage-SQLite%20%7C%20Postgres%20%7C%20MySQL-b45309)
 ![Replay Evidence](https://img.shields.io/badge/Replay-evidence%20driven-7c3aed)
 ![Tool Ledger](https://img.shields.io/badge/Tools-ledger%20guarded-d97706)
-AgentLedger `1.2.2` 是一个面向 AI Agent 的 agent execution safety、evidence 与 reliability layer。它不负责让 Agent 更会“思考”，而是让 Agent run 在 worker 崩溃、工具失败、prompt 变更和长任务恢复时，仍然具备持久化、可审计、可重放、可治理和可恢复能力。
+你的 Agent 调用了一个工具。它真的执行了吗？可以安全重试吗？之后能证明发生过什么吗？
+AgentLedger `1.2.4` 是面向 Agent Harness stack 的 runtime reliability layer。它不负责让 Agent 更会“思考”，也不替代完整 Harness 生态；它负责让 Agent run 在 worker 崩溃、工具失败、prompt 变更和长任务恢复时，仍然具备持久化、可审计、可重放、可治理和可恢复能力。
 大多数 Agent 框架关注 planning、reasoning 和 workflow logic。AgentLedger 放在 LangChain、LangGraph、CrewAI、AutoGen、OpenAI Agents SDK、LlamaIndex、Semantic Kernel 或自定义 Agent 的下方或旁边，提供 state、tool、evidence、replay、recovery 相关的 runtime guarantees。
+在完整 Harness stack 里，LangGraph、Temporal、Langfuse、MCP、model router、storage backend、sandbox provider 等系统可以各自负责擅长的层。AgentLedger 负责它们之间的 reliability substrate：durable execution、tool/model governance、evidence、replay、policy/sandbox boundary、cost/failure attribution 和 adapter contracts。
 Python 仍然是 reference implementation；Go、TypeScript、Rust 已有 native runtime-core baseline，并对齐同一份 language-neutral runtime contract。provider-specific driver 和 framework-native adapter 会按各语言生态差异处理。四语言实现对比和 adapter 边界见 `docs/zh/LANGUAGE_IMPLEMENTATION_COMPARISON.md`。
 ## 从这里开始
@@ -23,6 +27,11 @@ Python 仍然是 reference implementation；Go、TypeScript、Rust 已有 native
 | 安装并跑通第一个示例 | [docs/zh/GETTING_STARTED.md](docs/zh/GETTING_STARTED.md) |
 | 选择 Python / Go / TypeScript / Rust | [docs/zh/LANGUAGE_QUICKSTART.md](docs/zh/LANGUAGE_QUICKSTART.md) |
 | 找可运行示例 | [examples/README.md](examples/README.md)、[go/examples/README.md](go/examples/README.md)、[typescript/examples/README.md](typescript/examples/README.md)、[rust/examples/README.md](rust/examples/README.md) |
+| 查询 runtime 表 | [docs/zh/QUERY_EXAMPLES.md](docs/zh/QUERY_EXAMPLES.md) |
+| 理解 Harness stack 组合方式 | [docs/zh/HARNESS_STACK.md](docs/zh/HARNESS_STACK.md) |
+| 理解开源生态价值 | [docs/zh/OPEN_SOURCE_IMPACT.md](docs/zh/OPEN_SOURCE_IMPACT.md) |
+| 理解维护者职责 | [docs/zh/MAINTAINER_NOTES.md](docs/zh/MAINTAINER_NOTES.md) |
+| 规划 adoption 工作 | [docs/zh/ADOPTION.md](docs/zh/ADOPTION.md) |
 | 理解四语言哪些对齐、哪些不对齐 | [docs/zh/LANGUAGE_IMPLEMENTATION_COMPARISON.md](docs/zh/LANGUAGE_IMPLEMENTATION_COMPARISON.md) |
 | 安装 optional adapter packages | [docs/zh/ADAPTER_PACKAGING.md](docs/zh/ADAPTER_PACKAGING.md) |
 | 正确使用 Go | [go/README.md](go/README.md#install) |
@@ -72,6 +81,23 @@ Optional production adapter:
 例如 sandbox semantics 属于 core，但 sandbox infrastructure 不属于 core。core 负责 `SandboxPolicy`、fail-closed routing、audit/evidence record 和 replay safety；Docker、E2B、bubblewrap、Kubernetes/gVisor、Firecracker 或自定义 executor 都是 adapter。
+## Harness stack 组合方式
+AgentLedger 本身不是完整 Agent Harness。它的设计目标是和 Harness 生态里的其它系统组合：
+从本地最小 Harness 到 Temporal + LangGraph + observability 组合方式，见 [docs/zh/HARNESS_STACK.md](docs/zh/HARNESS_STACK.md)。
+| Harness 层 | 典型系统 | AgentLedger 角色 |
+| --- | --- | --- |
+| Agent workflow / planning | LangGraph、LangChain、CrewAI、AutoGen、OpenAI Agents SDK、自定义代码 | 用 durable state、policy、Tool Ledger、evidence 和 replay guarantees 包住 node 和 tool |
+| Durable orchestration | Temporal、Ray、Kubernetes workers | 在 worker step 内提供 agent-specific lease、fencing、checkpoint、cancellation、cost/failure attribution 和 replay semantics |
+| Observability / eval UI | Langfuse、LangSmith、OpenTelemetry、自定义 dashboard | 导出 structured runtime events、evidence bundle、trace/cost/failure data 和 correlation IDs |
+| Tool and context protocols | MCP、internal tool servers、provider SDK tools | 在副作用发生前强制执行 schema、permission、approval、sandbox、idempotency 和 audit |
+| Model providers / routers | OpenAI、Anthropic、Gemini、Bedrock、Ollama、LiteLLM、企业 gateway | 提供 runtime model-call contract、archived responses、budget/fallback/replay semantics 和 optional provider adapters |
+| Storage / artifacts | SQLite、Postgres、MySQL、S3/MinIO、内部存储 | 持久化 runtime metadata、state version、migration、blob ref 和 evidence ref，并通过 conformance 验证 |
+所以推荐的生产形态不是 `AgentLedger instead of LangGraph/Temporal/Langfuse`，而是 `AgentLedger with LangGraph/Temporal/Langfuse`：AgentLedger 负责治理这些系统自身难以强制保证的 model/tool/state boundary。
 ## 适合什么场景
 - 长期运行的 Agent task 需要在 crash 或 restart 后从最后一次 checkpoint 恢复
@@ -95,10 +121,12 @@ Optional production adapter:
 ## 示例
-仓库里既有最小 quickstart，也有一个更完整的多语言 Travel Assistant demo，用来展示 Python、Go、Rust、TypeScript 下相同的 runtime 思路。
+仓库里包含四语言 3-minute side-effect safety demo、MCP governance demo、最小 quickstart，以及一个更完整的多语言 Travel Assistant demo，用来展示 Python、Go、Rust、TypeScript 下相同的 runtime 思路。
-| 语言 | Demo | 运行方式 |
+| 目标 | Demo | 运行方式 |
 | --- | --- | --- |
+| 3 分钟副作用安全 demo | Python / Go / Rust / TypeScript | `PYTHONPATH=src python3 examples/three_minute_demo/demo.py`; `cd go && go run ./examples/three_minute_demo`; `cd rust && cargo run --example three_minute_demo`; `cd typescript && node examples/three_minute_demo/three_minute_demo.js` |
+| MCP tool governance | Python / Go / Rust / TypeScript | `PYTHONPATH=src python3 examples/mcp_governance/demo.py`; `cd go && go run ./examples/mcp_governance`; `cd rust && cargo run --example mcp_governance`; `cd typescript && node examples/mcp_governance/mcp_governance.js` |
 | Python | `examples/travel_assistant/demo.py` | `python3 examples/travel_assistant/demo.py` |
 | Go | `go/examples/travel_assistant/main.go` | `cd go && go run examples/travel_assistant/main.go` |
 | Rust | `rust/examples/travel_assistant.rs` | `cd rust && cargo run --example travel_assistant` |
@@ -253,7 +281,7 @@ AgentLedger 也不是新的 LLM SDK，不是 workflow engine，不是通用 obse
 ## 当前成熟度
-AgentLedger 1.2.0 是 stable runtime-core release，Python 是 reference implementation，Go、TypeScript、Rust 已由共享 runtime-core parity gate 覆盖；适合本地使用、framework adapter integration、reliability semantics 验证，以及在明确 adapter 边界下做 production pilot 准备。
+AgentLedger 1.2.4 是 stable runtime-core release，Python 是 reference implementation，Go、TypeScript、Rust 已由共享 runtime-core parity gate 覆盖；适合本地使用、framework adapter integration、reliability semantics 验证，以及在明确 adapter 边界下做 production pilot 准备。
 runtime-core contract 已稳定；optional production adapter 和外部基础设施加固仍按独立阶段推进。详见 [docs/MATURITY_MODEL.md](docs/MATURITY_MODEL.md)、[docs/zh/IMPLEMENTATION_STATUS.md](docs/zh/IMPLEMENTATION_STATUS.md) 和 [docs/zh/ROADMAP.md](docs/zh/ROADMAP.md)。

{agentledger_runtime-1.2.2 → agentledger_runtime-1.2.4}/contracts/conformance/optional_adapters.v1.json RENAMED Viewed

@@ -6,8 +6,8 @@
     {
       "name": "optional_backend_capabilities_are_discoverable",
       "requires": ["optional_adapter_capabilities"],
-      "expected_categories": ["storage", "blobstore", "sandbox", "framework", "mcp", "shadow"],
-      "expected_adapters": ["postgres", "mysql", "s3", "docker", "e2b", "bubblewrap", "kubernetes", "gvisor", "firecracker", "langgraph", "langchain", "crewai", "autogen", "openai-agents-sdk", "llamaindex", "semantic-kernel", "mcp-transport", "shadow-runner"],
+      "expected_categories": ["storage", "blobstore", "sandbox", "framework", "mcp", "observability", "shadow"],
+      "expected_adapters": ["postgres", "mysql", "s3", "docker", "e2b", "bubblewrap", "kubernetes", "gvisor", "firecracker", "langgraph", "langchain", "crewai", "autogen", "openai-agents-sdk", "llamaindex", "semantic-kernel", "mcp-transport", "langfuse", "shadow-runner"],
       "expected_contract": {
         "core_imports_heavy_sdks": false,
         "adapter_is_optional": true,
@@ -16,7 +16,7 @@
     }
   ],
   "non_goals": [
-    "This fixture does not require live Postgres, MySQL, S3, Docker, Kubernetes, E2B, MCP, LangGraph, LangChain, CrewAI, AutoGen, LlamaIndex, or Semantic Kernel dependencies.",
+    "This fixture does not require live Postgres, MySQL, S3, Docker, Kubernetes, E2B, MCP, LangGraph, LangChain, CrewAI, AutoGen, LlamaIndex, Semantic Kernel, or Langfuse dependencies.",
     "Concrete production adapters may live in optional packages or be injected by the application, while runtime core owns contracts, policy, audit, evidence, and fail-closed behavior."
   ]
 }

{agentledger_runtime-1.2.2 → agentledger_runtime-1.2.4}/docs/ADAPTER_CERTIFICATION.md RENAMED Viewed

@@ -212,10 +212,10 @@ This does not make an adapter production-ready by itself. It gives users a concr
 AgentLedger can generate a machine-readable starting bundle for official adapter profiles:
 ```bash
-PYTHONPATH=src python3 -m agentledger adapter certify --kind postgres --adapter-version 1.2.2
-PYTHONPATH=src python3 -m agentledger adapter certify --kind mysql --adapter-version 1.2.2 --out ./mysql-certification.json
-PYTHONPATH=src python3 -m agentledger adapter certify --kind s3 --adapter-version 1.2.2 --out ./s3-certification.json
-PYTHONPATH=src python3 -m agentledger adapter certify --kind langgraph --adapter-version 1.2.2 --package-name agentledger-langgraph
+PYTHONPATH=src python3 -m agentledger adapter certify --kind postgres --adapter-version 1.2.4
+PYTHONPATH=src python3 -m agentledger adapter certify --kind mysql --adapter-version 1.2.4 --out ./mysql-certification.json
+PYTHONPATH=src python3 -m agentledger adapter certify --kind s3 --adapter-version 1.2.4 --out ./s3-certification.json
+PYTHONPATH=src python3 -m agentledger adapter certify --kind langgraph --adapter-version 1.2.4 --package-name agentledger-langgraph
 ```
 Supported built-in profiles:

{agentledger_runtime-1.2.2 → agentledger_runtime-1.2.4}/docs/ADAPTER_PACKAGING.md RENAMED Viewed

@@ -1,6 +1,6 @@
 # Adapter Packaging
-AgentLedger `1.2.1` introduced the adapter packaging model, and `1.2.2` extends that model with the official MySQL storage adapter boundary. Runtime-core stays small and dependency-light; concrete integrations move into optional adapter packages that can be installed through extras or directly.
+AgentLedger `1.2.1` introduced the adapter packaging model, `1.2.2` extended it with the official MySQL storage adapter boundary, and `1.2.3` adds a dependency-free Langfuse evidence/trace export boundary. Runtime-core stays small and dependency-light; concrete integrations move into optional adapter packages that can be installed through extras or directly.
 ## Why Split Adapters
@@ -48,6 +48,7 @@ pip install "agentledger-runtime[s3]"
 pip install "agentledger-runtime[langgraph]"
 pip install "agentledger-runtime[mcp]"
 pip install "agentledger-runtime[otel]"
+pip install "agentledger-runtime[langfuse]"
 pip install "agentledger-runtime[docker]"
 pip install "agentledger-runtime[all]"
 ```
@@ -61,6 +62,7 @@ pip install agentledger-s3
 pip install agentledger-langgraph
 pip install agentledger-mcp
 pip install agentledger-otel
+pip install agentledger-langfuse
 pip install agentledger-sandbox-docker
 ```
@@ -85,6 +87,7 @@ agentledger-runtime/
     agentledger-langgraph/
     agentledger-mcp/
     agentledger-otel/
+    agentledger-langfuse/
     agentledger-sandbox-docker/
   typescript/
     src/adapters/                        # runtime subpath exports
@@ -95,6 +98,7 @@ agentledger-runtime/
       agentledger-langgraph/
       agentledger-mcp/                   # npm package name: agentledger-mcp-adapter
       agentledger-otel/
+      agentledger-langfuse/
       agentledger-sandbox-docker/
   go/adapters/
     postgres/
@@ -102,6 +106,7 @@ agentledger-runtime/
     s3/
     mcp/
     otel/
+    langfuse/
     sandbox/docker/
     framework/
   rust/
@@ -111,6 +116,7 @@ agentledger-runtime/
       agentledger-s3/
       agentledger-mcp/
       agentledger-otel/
+      agentledger-langfuse/
       agentledger-sandbox-docker/
       agentledger-framework/
 ```
@@ -134,6 +140,7 @@ Each adapter package should provide:
 | `agentledger-langgraph` | LangGraph checkpointer/node wrappers around the dependency-free facade | Core facade is dependency-free; optional native SDK use belongs behind package extras or follow-up smoke matrices. |
 | `agentledger-mcp` / `agentledger-mcp-adapter` on npm | MCP-style tool/context mapping package boundary | Current package is dependency-light; exact MCP SDK client/server transport is a follow-up adapter hardening item. |
 | `agentledger-otel` | OTLP JSON/export package boundary around AgentLedger spans | Current package is dependency-light; hardened OpenTelemetry SDK wiring is follow-up work. |
+| `agentledger-langfuse` | Langfuse-style evidence/trace payload export | Current package is dependency-light; Langfuse SDK/server ingestion behavior remains application/deployment validation. |
 | `agentledger-sandbox-docker` | Docker sandbox executor package and local/team recipes | Current boundary can use Docker CLI/manifest semantics; daemon hardening, network policy, and resource validation are external. |
 Language fit matters. `agentledger-langgraph` is first-class for Python and TypeScript/Node because those ecosystems have LangGraph packages. Go and Rust expose a generic `framework` adapter boundary instead of pretending a native LangGraph ecosystem exists there.
@@ -194,6 +201,7 @@ s3 = ["agentledger-s3>=1.2,<2"]
 langgraph = ["agentledger-langgraph>=1.2,<2"]
 mcp = ["agentledger-mcp>=1.2,<2"]
 otel = ["agentledger-otel>=1.2,<2"]
+langfuse = ["agentledger-langfuse>=1.2,<2"]
 docker = ["agentledger-sandbox-docker>=1.2,<2"]
 all = [
   "agentledger-postgres>=1.2,<2",
@@ -202,6 +210,7 @@ all = [
   "agentledger-langgraph>=1.2,<2",
   "agentledger-mcp>=1.2,<2",
   "agentledger-otel>=1.2,<2",
+  "agentledger-langfuse>=1.2,<2",
   "agentledger-sandbox-docker>=1.2,<2",
 ]
 ```
@@ -214,8 +223,8 @@ The `1.2.x` packaging release is expected to pass:
 ```bash
 PYTHONDONTWRITEBYTECODE=1 PYTHONPATH=src python3 -m unittest discover -s tests -q
-PYTHONDONTWRITEBYTECODE=1 PYTHONPATH=src python3 -m agentledger adapter certify --kind postgres --adapter-version 1.2.2
-PYTHONDONTWRITEBYTECODE=1 PYTHONPATH=src python3 -m agentledger adapter certify --kind mysql --adapter-version 1.2.2
+PYTHONDONTWRITEBYTECODE=1 PYTHONPATH=src python3 -m agentledger adapter certify --kind postgres --adapter-version 1.2.4
+PYTHONDONTWRITEBYTECODE=1 PYTHONPATH=src python3 -m agentledger adapter certify --kind mysql --adapter-version 1.2.4
 PYTHONDONTWRITEBYTECODE=1 PYTHONPATH=src python3 scripts/check_adapter_packages.py
 go test ./...
 cd typescript && npm test

{agentledger_runtime-1.2.2 → agentledger_runtime-1.2.4}/docs/ADAPTER_ROADMAP.md RENAMED Viewed

@@ -34,6 +34,7 @@ These should be first-class optional packages because they match common producti
 | Framework | LangGraph | High overlap with stateful agent workflows; AgentLedger adds Tool Ledger, evidence, replay, policy, and adapter certification. | `framework_adapters.v1.json`, checkpoint boundary, optional capability descriptor | `agentledger-langgraph` |
 | Tool/context protocol | MCP transport | MCP is a natural tool/context boundary for agents. Runtime should govern MCP tools without owning every tool server. | `mcp_adapters.v1.json`, `optional_adapters.v1.json` | `agentledger-mcp` / `agentledger-mcp-adapter` on npm |
 | Observability | OpenTelemetry exporter/transport | Enterprises need traces in existing observability stacks. | `otlp_trace_export.v1.json` | `agentledger-otel` |
+| Observability | Langfuse exporter | Langfuse is a common LLM observability surface; AgentLedger should export evidence and trace payloads without replacing Langfuse. | evidence/trace export, correlation IDs | `agentledger-langfuse` |
 | Sandbox | Docker sandbox | Common local/team isolation backend; useful before Kubernetes/gVisor/Firecracker. | `policy_approval_sandbox.v1.json`, sandbox policy/result boundary | `agentledger-sandbox-docker` |
 | Scheduler/backend | Temporal bridge | Temporal can own durable workflow orchestration while AgentLedger owns agent-specific evidence/tool/state semantics. | `scheduler.v1.json`, execution backend boundary | `agentledger-temporal` |
@@ -55,6 +56,7 @@ These are valuable but should follow the Priority 1 adapters or remain thinner f
 | Sandbox | E2B | Good managed remote sandbox for code/tool execution. | Keep as optional remote executor adapter. |
 | Distributed execution | Ray bridge | Useful for Python distributed worker pools. | Ray should own cluster scheduling; AgentLedger owns run semantics. |
 | Deployment | Kubernetes worker recipe | Useful for pilots. | Recipe/Helm/examples first; full platform later only if needed. |
+| Model providers/router | OpenAI, Anthropic, Gemini, Bedrock, Azure OpenAI, Ollama, LiteLLM-style bridge | Model calls affect evidence, cost, replay, fallback, and budget enforcement, but provider SDKs should not be runtime-core dependencies. | Wait for `ModelGateway`/`ModelRouter` contract before stabilizing provider packages. |
 ## Priority 3: Experimental Or Community Adapters
@@ -86,6 +88,7 @@ Python remains the reference implementation, but official adapters should conver
 | Kubernetes sandbox/backend | Recommended | Recommended | Recommended | Recommended | Prefer manifest/dry-run contract plus optional execution. |
 | Temporal bridge | Recommended | Required when Go runtime matures | Recommended | Community/optional | Match Temporal ecosystem strength per language. |
 | OpenTelemetry | Required official | Required official | Required official | Required official | Standard enterprise observability path. |
+| Model provider/router | Recommended after ModelGateway contract | Recommended after ModelGateway contract | Recommended after ModelGateway contract | Recommended after ModelGateway contract | Keep provider SDKs optional; use injected clients and conformance fixtures. |
 ## Non-Negotiable Adapter Requirements
@@ -109,3 +112,4 @@ Every official adapter must provide:
 6. LangChain / CrewAI / AutoGen / OpenAI Agents SDK / LlamaIndex / Semantic Kernel facades where ecosystems are stable.
 7. Kubernetes sandbox/backend recipe, then optional execution adapter.
 8. Temporal/Ray/Kubernetes scheduler/backend bridges based on real user demand.
+9. Model provider/router adapters after the runtime `ModelGateway` contract is stable.

agentledger_runtime-1.2.4/docs/ADOPTION.md ADDED Viewed

@@ -0,0 +1,58 @@
+# Adoption Plan
+This document tracks near-term work that makes AgentLedger easier to understand, try, discuss, and validate. It does not change runtime-core scope.
+## Why This Track Exists
+AgentLedger already has substantial runtime-core depth. The next adoption bottleneck is not another large feature; it is helping a new user answer three questions quickly:
+```text
+What problem does this solve?
+Can I see it in three minutes?
+Can I verify the reliability claim myself?
+```
+## Current Adoption Priorities
+| Priority | Item | Acceptance criteria |
+|---|---|---|
+| P0 | 3-minute demo | A user can run the Python, Go, TypeScript, or Rust 3-minute demo and see one external side effect, one Tool Ledger entry, safe retry, and replay-safe evidence. |
+| P0 | README first-screen pain point | The first screen explains the tool side-effect problem before listing architecture. |
+| P1 | MCP governance example | Dependency-free MCP-style tools demonstrate schema, approval, sandbox metadata, idempotency, and audit evidence across Python, Go, TypeScript, and Rust. |
+| P1 | Public issue/discussion list | The next adoption tasks are easy to open as GitHub issues or discussions. |
+| P1 | Case study template | A real integration can be documented without exposing private data. |
+## Suggested Public Issues
+These can be opened as GitHub issues when the project is ready for public tracking.
+| Title | Type | Why it matters |
+|---|---|---|
+| Build a cross-language 3-minute Tool Ledger demo | example | Shows the core value in a few minutes across Python, Go, TypeScript, and Rust. |
+| Add OpenAI Agents SDK approval/replay example | example | Connects AgentLedger to a major agent SDK boundary without claiming official endorsement. |
+| Add MCP tool governance example | example | Shows how MCP-style tools should pass through policy, approval, sandbox, idempotency, and evidence. |
+| Prototype AgentLedger Inspector | product | Makes run timelines, Tool Ledger, approvals, replay, and failures visible. |
+| Add Temporal bridge example | integration | Clarifies Temporal owns workflow lifecycle while AgentLedger owns node-internal reliability. |
+| Add tool-injection risk scanner | security | Detects risky tool schemas, missing approval/sandbox, and runtime-boundary bypass patterns. |
+| Publish legal-agent case study | case study | Shows a realistic audit/evidence use case without private data. |
+## Case Study Rules
+Case studies should be useful but conservative:
+- remove private data, customer data, secrets, and internal implementation details
+- describe the runtime problem before describing AgentLedger
+- show the AgentLedger integration boundary
+- include concrete evidence artifacts only when safe
+- avoid claiming production hardening unless real operational evidence exists
+## Good Adoption Evidence
+- runnable examples
+- short terminal recordings or GIFs
+- public issues and discussions
+- package downloads
+- real integration notes
+- external demos or blog posts
+- adapter conformance reports
+- real-service hardening reports

{agentledger_runtime-1.2.2 → agentledger_runtime-1.2.4}/docs/ARCHITECTURE.md RENAMED Viewed

@@ -1,9 +1,11 @@
 # Architecture
-AgentLedger is an agent execution safety and evidence layer. It is not an agent framework, model SDK, workflow engine, eval system, observability suite, RAG system, or sandbox infrastructure provider.
+AgentLedger is a runtime reliability layer for Agent Harness stacks. It is not an agent framework, model SDK, workflow engine, eval system, observability suite, RAG system, or sandbox infrastructure provider.
 The architecture keeps runtime core thin but hard to replace: it only owns guarantees that must be enforced at the boundary between agent logic and external effects. Mature systems integrate through adapters and conformance tests instead of being rebuilt inside core.
+In a complete harness stack, AgentLedger is the reliability substrate under or beside systems such as LangGraph, Temporal, Langfuse, MCP, model providers, storage backends, and sandbox providers.
 The practical split is:
 ```text

{agentledger_runtime-1.2.2 → agentledger_runtime-1.2.4}/docs/COMPLETE_CORE_PARITY_CHECKLIST.md RENAMED Viewed

@@ -55,8 +55,8 @@ Excluded or not applicable:
 | CLI baseline | `--help`, `doctor`, `version`, `quickstart`, `conformance`, `contract validate`, `contract export` | `agentledger` CLI | `agentledger-go` | `agentledger-ts` | `agentledger-rust` | `scripts/check_complete_core_parity.py` | done |
 | Quickstart docs | Language quickstart and adapter quickstart | README + docs | `go/README.md` | `typescript/README.md` | `rust/README.md` | docs link check | done |
 | Examples | Runnable examples for core runtime and adapter API | Python examples | `go/examples/quickstart/main.go` | `typescript/examples/quickstart/quickstart.js` | `rust/examples/quickstart.rs` | `scripts/check_complete_core_parity.py` | done |
-| Package metadata | Installable package metadata in the current release train | PyPI metadata `1.2.2` | `go.mod` tag `go/v1.2.2` | npm package `1.2.2` | `Cargo.toml` `1.2.2` | `scripts/check_complete_core_parity.py` version check | done |
-| Package install smoke | Install/import/run from published or built package | PyPI `agentledger-runtime==1.2.2` release smoke after publish | `go/v1.2.2` clean external `go get` smoke after tag | npm `agentledger-runtime@1.2.2` release smoke after publish | crates.io `agentledger-runtime==1.2.2` release smoke after publish | publish logs + `scripts/check_complete_core_parity.py` dry-run gates | release-gated |
+| Package metadata | Installable package metadata in the current release train | PyPI metadata `1.2.4` | `go.mod` tag `go/v1.2.4` | npm package `1.2.4` | `Cargo.toml` `1.2.4` | `scripts/check_complete_core_parity.py` version check | done |
+| Package install smoke | Install/import/run from published or built package | PyPI `agentledger-runtime==1.2.4` release smoke after publish | `go/v1.2.4` clean external `go get` smoke after tag | npm `agentledger-runtime@1.2.4` release smoke after publish | crates.io `agentledger-runtime==1.2.4` release smoke after publish | publish logs + `scripts/check_complete_core_parity.py` dry-run gates | release-gated |
 | Release docs | Changelog, release checklist, status docs reflect exact parity boundary | docs | docs | docs | docs | docs link check | done |
 | Comparison docs | Honest overlap with LangChain/LangGraph/LangSmith/Langfuse/etc. | docs | docs | docs | docs | docs link check | done |
@@ -67,7 +67,7 @@ The strict core parity gate now covers the previously weak areas:
 1. CLI baseline is automatically checked for all four language CLIs.
 2. Go, TypeScript, and Rust have runnable quickstart example files.
 3. TypeScript and Rust package surfaces have package dry-run checks (`npm pack --dry-run`, `cargo package --allow-dirty --no-verify`) and post-publish install smokes are required for release.
-4. Package versions are checked across Python, TypeScript, and Rust; the current release train is aligned at `1.2.2`.
+4. Package versions are checked across Python, TypeScript, and Rust; the current release train is aligned at `1.2.4`.
 5. Go external module consumption is tag/release dependent; each release should repeat the clean external `go get` smoke after pushing the `go/vX.Y.Z` tag.
 6. Real service-backed hardening remains out of scope for core parity and stays documented as optional follow-up work.

{agentledger_runtime-1.2.2 → agentledger_runtime-1.2.4}/docs/EXECUTION_BACKENDS.md RENAMED Viewed

@@ -160,7 +160,7 @@ LangGraph
   agent graph / workflow logic / multi-agent orchestration
 AgentLedger
-  inner agent execution safety, evidence, and tool governance layer
+  inner runtime reliability, evidence, and tool/model governance layer
 ```
 A typical stack:

{agentledger_runtime-1.2.2 → agentledger_runtime-1.2.4}/docs/EXTENSIBILITY.md RENAMED Viewed

@@ -364,7 +364,7 @@ Blob adapters should implement `BlobStoreProtocol` and pass `BlobStoreConformanc
 ## Observability Adapter Boundary
-`TraceExporter` emits dependency-free JSONL spans from evidence events. `OTLPTraceExporter` translates the same spans into OTLP/JSON without importing OpenTelemetry SDKs and can optionally POST that JSON to a configured collector. Observability adapters should preserve this format for OpenTelemetry collectors, LangSmith-style backends, or custom trace stores without changing runtime-core event semantics.
+`TraceExporter` emits dependency-free JSONL spans from evidence events. `OTLPTraceExporter` translates the same spans into OTLP/JSON without importing OpenTelemetry SDKs and can optionally POST that JSON to a configured collector. `agentledger-langfuse` converts the same evidence spans into a Langfuse-style ingestion payload. Observability adapters should preserve this format for OpenTelemetry collectors, Langfuse/LangSmith-style backends, or custom trace stores without changing runtime-core event semantics.
 `TimeTravelDebugger` is also event-log based. It reconstructs committed state by applying initial state and committed JSON merge patches in event sequence order. Debugging adapters or UIs should preserve this read-only, side-effect-free model.

agentledger-runtime 1.2.2__tar.gz → 1.2.4__tar.gz

agentledger-runtime 1.2.2tar.gz → 1.2.4tar.gz