[danipl]

Releases the Python SDK 1.24.0, the SDK version for Infrahub 1.12.0 (IFC-3589). Until now, only the betas v1.24.0b0 and v1.24.0b1 were published.

This pull request brings infrahub-release-1.12 (commit 1b19c283) into stable, as the 1.23.0 release did in #1259. It adds one commit to the release branch:

• Adds the 1.24.0 section to CHANGELOG.md, built by towncrier from 58 changelog fragments.
• Deletes those 58 fragments.
• Deletes 14 more fragments without listing them under 1.24.0. The 1.23.0 release had already used them on stable (commit c7313d54), but infrahub-release-1.12 still had copies with identical text.

Merge with a merge commit

Merge this pull request with a merge commit, not a squash. A squash would break the shared history between stable and infrahub-release-1.12.

Next steps after the merge

• Tag the merge commit as v1.24.0 and publish the GitHub release. The release workflow then publishes the package to PyPI.
• Bring the changelog commit back into infrahub-release-1.12 and infrahub-develop.

🤖 Generated with Claude Code

[cloudflare-workers-and-pages]

Deploying infrahub-sdk-python with    Cloudflare Pages

Latest commit:	53cdfdd
Status:	✅  Deploy successful!
Preview URL:	https://cbce2007.infrahub-sdk-python.pages.dev
Branch Preview URL:	https://dp-20261109-release-sdk-1-24.infrahub-sdk-python.pages.dev

View logs

[codecov]

Codecov Report

❌ Patch coverage is 97.50701% with 80 lines in your changes missing coverage. Please review.

Files with missing lines	Patch %	Lines
infrahub_sdk/query_groups.py	89.91%	6 Missing and 6 partials ⚠️
infrahub_sdk/exceptions/base.py	95.52%	7 Missing and 4 partials ⚠️
infrahub_sdk/protocols_generator/generator.py	76.31%	6 Missing and 3 partials ⚠️
infrahub_sdk/ctl/utils.py	84.90%	6 Missing and 2 partials ⚠️
infrahub_sdk/node/relationship.py	93.25%	1 Missing and 5 partials ⚠️
infrahub_sdk/protocols_base.py	25.00%	0 Missing and 6 partials ⚠️
infrahub_sdk/ctl/cli_commands.py	42.85%	4 Missing ⚠️
infrahub_sdk/ctl/troubleshooting.py	93.33%	3 Missing and 1 partial ⚠️
infrahub_sdk/spec/object.py	77.77%	3 Missing and 1 partial ⚠️
infrahub_sdk/testing/repository.py	86.36%	1 Missing and 2 partials ⚠️
... and 8 more

@@            Coverage Diff             @@
##           stable    #1503      +/-   ##
==========================================
- Coverage   85.18%   83.78%   -1.41%     
==========================================
  Files         148      157       +9     
  Lines       13368    15656    +2288     
  Branches     1975     2114     +139     
==========================================
+ Hits        11388    13117    +1729     
- Misses       1412     1979     +567     
+ Partials      568      560       -8     

Flag	Coverage Δ
integration-tests	?
python-3.10	62.91% <80.95%> (+4.58%)	⬆️
python-3.11	62.91% <80.95%> (+4.58%)	⬆️
python-3.12	62.90% <80.95%> (+4.56%)	⬆️
python-3.13	62.91% <80.95%> (+4.57%)	⬆️
python-3.14	62.91% <80.95%> (+4.57%)	⬆️
python-3.15	62.91% <80.95%> (+4.57%)	⬆️
python-filler-3.12	23.03% <16.57%> (-0.67%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

Files with missing lines	Coverage Δ
infrahub_sdk/constants.py	94.44% <100.00%> (+1.11%)	⬆️
infrahub_sdk/ctl/object/create.py	93.18% <100.00%> (-4.44%)	⬇️
infrahub_sdk/ctl/object/inline_peers.py	100.00% <100.00%> (ø)
infrahub_sdk/exceptions/__init__.py	100.00% <100.00%> (ø)
infrahub_sdk/exceptions/catalogue.py	100.00% <100.00%> (ø)
infrahub_sdk/file_handler.py	87.56% <100.00%> (ø)
infrahub_sdk/graph_traversal/query.py	100.00% <100.00%> (ø)
infrahub_sdk/node/__init__.py	100.00% <100.00%> (ø)
infrahub_sdk/node/_only.py	100.00% <100.00%> (ø)
infrahub_sdk/node/attribute.py	100.00% <100.00%> (ø)
... and 32 more

... and 18 files with indirect coverage changes

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[cubic-dev-ai]

20 issues found across 181 files

Prompt for AI agents (unresolved issues)

Check if these issues are valid — if so, understand the root cause of each and fix them. When an issue isn't valid or won't be fixed in this PR, reply in its thread with the reason and then resolve the thread. If appropriate, use sub-agents to investigate and fix each issue separately.


<file name="pyproject.toml">

<violation number="1" location="pyproject.toml:69">
P2: Custom agent: **Detect conflicting package versions across dependency files**

This adds a second `pytest` specifier in `pyproject.toml`: the `testing` extra allows `pytest>=7.0`, while the `tests` group requires `pytest>=9.0,<9.1`. Align the specifiers to satisfy the dependency-version consistency rule.</violation>
</file>

<file name="infrahub_sdk/transfer/importer/json.py">

<violation number="1" location="infrahub_sdk/transfer/importer/json.py:199">
P2: Custom agent: **Flag AI Slop and Fabricated Changes**

This `continue_on_error` rendering change has no regression coverage: the importer tests never enable `continue_on_error` or exercise `execute_batches`. Add assertions for empty `GraphQLError.errors` and entries without `message` so both fallback behaviors stay protected.</violation>
</file>

<file name="docs/docs/python-sdk/sdk_ref/infrahub_sdk/node/related_node.mdx">

<violation number="1" location="docs/docs/python-sdk/sdk_ref/infrahub_sdk/node/related_node.mdx:129">
P3: The new `is_loaded` Returns text is split into separate bullets, so the sentence reads as fragments when rendered. The continuation lines should be plain text, not `- ` items. This comes from the doc generator, so the fix may belong there rather than in this file.</violation>
</file>

<file name=".github/workflows/ci.yml">

<violation number="1" location=".github/workflows/ci.yml:359">
P2: The install matrix omits Python 3.15 even though the project supports it, so this new plain-wheel install check does not cover every supported interpreter. Add 3.15 to the matrix.</violation>
</file>

<file name="infrahub_sdk/exceptions/base.py">

<violation number="1" location="infrahub_sdk/exceptions/base.py:102">
P3: An explicit empty `message` is replaced by the generated default here. Use an `is not None` check, as `GraphQLError` does, so a deliberate empty message is kept. The same pattern appears in the other optional-message constructors in this file.</violation>

<violation number="2" location="infrahub_sdk/exceptions/base.py:209">
P2: Pickling and unpickling these three exceptions fails with a TypeError because their constructors take more than the message. Add a `__reduce__` that returns `(_rebuild_error, (type(self), self.args, self.__dict__))`, as `FieldNotLoadedError` already does, to `RateLimitError`, `VersionNotSupportedError`, and `ObjectValidationError`. Round-trip coverage in tests/unit/sdk/test_exceptions.py does not include them.</violation>

<violation number="3" location="infrahub_sdk/exceptions/base.py:456">
P3: This docstring is copied from `UninitializedError`, so it describes the wrong failure. Replace it with a sentence about an invalid server response.</violation>
</file>

<file name="infrahub_sdk/trace_context.py">

<violation number="1" location="infrahub_sdk/trace_context.py:19">
P2: When OTEL_PROPAGATORS is invalid, this import fails on every request, because the failed module is not cached and the `ValueError` path only caches the warning. Cache the failure result too, so the import is not re-attempted per request.</violation>
</file>

<file name="AGENTS.md">

<violation number="1" location="AGENTS.md:65">
P3: The new Branches section says Infrahub pins `python_sdk` to the matching branch, but the release flow in this PR moves the `infrahub-release-1.12` pin to the `v1.24.0` tag. Add a note that pins move to the release tag after a final release.</violation>

<violation number="2" location="AGENTS.md:65">
P3: `python_sdk` is a git submodule, so Infrahub pins a commit rather than tracking an SDK branch; this wording can make readers expect branch-following behavior. Say that the submodule points to a commit from the matching SDK branch.</violation>
</file>

<file name="infrahub_sdk/ctl/cli_commands.py">

<violation number="1" location="infrahub_sdk/ctl/cli_commands.py:247">
P3: The comment above this `except` is inaccurate. The GraphQLError clause below already prints the message when `errors` is empty, so that is not why these lookup misses are re-raised. The real reason is that `catch_exception` renders them with the `Error:` prefix, which the tests check. Rewrite the comment so a maintainer does not remove the branch on the belief that it is redundant.</violation>
</file>

<file name="docs/docs/python-sdk/topics/error_handling.mdx">

<violation number="1" location="docs/docs/python-sdk/topics/error_handling.mdx:196">
P3: This claims every fallback is logged at debug level. On the authentication path, a response with a code that has no class raises `AuthenticationError` without a log line, because the logging only runs when no code was resolved. Narrow the claim to the cases that actually log.</violation>
</file>

<file name="infrahub_sdk/node/attribute.py">

<violation number="1" location="infrahub_sdk/node/attribute.py:220">
P2: `detach_from_pool()` sets `_has_data = True`, which marks the attribute as loaded even when its value was never fetched. A later `.value` read on a node from `only` then returns `None` silently instead of raising `FieldNotLoadedError`. The detach only needs to send `from_pool: None`, so drop the `_has_data` assignment here.</violation>
</file>

<file name="infrahub_sdk/yaml.py">

<violation number="1" location="infrahub_sdk/yaml.py:180">
P2: Errors from `spec` models are reported with a path relative to `spec`, but the message omits the `spec` prefix. With a file whose top-level `kind: Object` is valid, `spec` missing `kind` reads as `kind | Field required (missing)`, which looks like the top-level field is absent. Prefix the spec-level location, for example by passing ("spec",) into `_content_error` from the `spec` callers in `spec/object.py` and `spec/menu.py`.</violation>
</file>

<file name="CHANGELOG.md">

<violation number="1" location="CHANGELOG.md:54">
P3: This release note claims to list all new catalogue exception classes but omits the exported `WorkerTimeoutError`. Add `WorkerTimeoutError` to the list so users can discover the new `WORKER_TIMEOUT` exception.</violation>
</file>

<file name="docs/docs/python-sdk/guides/installation.mdx">

<violation number="1" location="docs/docs/python-sdk/guides/installation.mdx:36">
P2: This description points users to the wrong file: the plugin reads `.infrahub.yml` as repository configuration and discovers `infrahub_tests` only in `test_*.yml`/`test_*.yaml` files. Document the test-file naming and location instead.</violation>
</file>

<file name="infrahub_sdk/query_groups.py">

<violation number="1" location="infrahub_sdk/query_groups.py:37">
P2: `WORKER_TIMEOUT` is missing from `_REQUEST_FAILURE_CODES`, so a server-side worker timeout is recorded as a refusal of the current member and cleanup continues against the remaining members. Add it to the request-level codes so the timeout stops the reap and is re-raised unchanged.</violation>
</file>

<file name="infrahub_sdk/testing/repository.py">

<violation number="1" location="infrahub_sdk/testing/repository.py:91">
P2: `core.sharedRepository` does not make future checked-out worktree files world-writable, so files added after `_share()` runs can revert to the container user's umask and remain unwritable by another Infrahub container user. Apply the shared mode whenever the worktree is updated, or use a mechanism that enforces shared permissions for future worktree files.</violation>
</file>

<file name="infrahub_sdk/ctl/troubleshooting.py">

<violation number="1" location="infrahub_sdk/ctl/troubleshooting.py:74">
P3: These new troubleshooting commands cross the Infrahub server boundary, but coverage only verifies mocked HTTP responses. Add a functional or integration test against a compatible Infrahub server to verify the endpoint contract and response handling.

(Based on your team's feedback about requiring real-server coverage for server-bound SDK and CLI behavior.) [ID: 8542b3a4-57c4-4f1c-abe8-ef576231d2c1]</violation>
</file>

<file name="docs/docs/python-sdk/sdk_ref/infrahub_sdk/node/inline_peers.mdx">

<violation number="1" location="docs/docs/python-sdk/sdk_ref/infrahub_sdk/node/inline_peers.mdx:2">
P3: This SDK reference exposes the internal `peers_are_inline_input` predicate as a public API. Remove this page, or document the supported `RelationshipManager` behavior instead; keep internal implementation helpers out of the SDK reference.</violation>
</file>

<sub>Reply with feedback, questions, or to request a fix.

View guided diff | Turn on auto-fix | Re-trigger cubic</sub>

[cubic-dev-ai]

P2: Custom agent: Detect conflicting package versions across dependency files

This adds a second pytest specifier in pyproject.toml: the testing extra allows pytest>=7.0, while the tests group requires pytest>=9.0,<9.1. Align the specifiers to satisfy the dependency-version consistency rule.

Prompt for AI agents

Check if this issue is valid — if so, understand the root cause and fix it. When an issue isn't valid or won't be fixed in this PR, reply in its thread with the reason and then resolve the thread. At pyproject.toml, line 69:

<comment>This adds a second `pytest` specifier in `pyproject.toml`: the `testing` extra allows `pytest>=7.0`, while the `tests` group requires `pytest>=9.0,<9.1`. Align the specifiers to satisfy the dependency-version consistency rule.</comment>

<file context>
@@ -43,38 +52,48 @@ infrahubctl = "infrahub_sdk.ctl.cli:app"
+# The bundled `pytest-infrahub` plugin and the schemas under `infrahub_sdk.testing`, for writing
+# tests against your own Transforms, Queries and Checks.
+testing = [
+    "pytest>=7.0",
+]
+
</file context>

Suggested change

	"pytest>=7.0",
	"pytest>=9.0,<9.1",

[cubic-dev-ai]

P2: Custom agent: Flag AI Slop and Fabricated Changes

This continue_on_error rendering change has no regression coverage: the importer tests never enable continue_on_error or exercise execute_batches. Add assertions for empty GraphQLError.errors and entries without message so both fallback behaviors stay protected.

Prompt for AI agents

Check if this issue is valid — if so, understand the root cause and fix it. When an issue isn't valid or won't be fixed in this PR, reply in its thread with the reason and then resolve the thread. At infrahub_sdk/transfer/importer/json.py, line 199:

<comment>This `continue_on_error` rendering change has no regression coverage: the importer tests never enable `continue_on_error` or exercise `execute_batches`. Add assertions for empty `GraphQLError.errors` and entries without `message` so both fallback behaviors stay protected.</comment>

<file context>
@@ -194,9 +194,13 @@ async def execute_batches(
-                    if isinstance(result, GraphQLError):
+                    # Keyed on there being server errors to render rather than on the class: a
+                    # lookup miss is a GraphQLError too, and carries its reason in its message.
+                    if isinstance(result, GraphQLError) and result.errors:
                         error_name = type(result).__name__
-                        error_msgs = [err["message"] for err in result.errors]
</file context>

[cubic-dev-ai]

P2: The install matrix omits Python 3.15 even though the project supports it, so this new plain-wheel install check does not cover every supported interpreter. Add 3.15 to the matrix.

Prompt for AI agents

Check if this issue is valid — if so, understand the root cause and fix it. When an issue isn't valid or won't be fixed in this PR, reply in its thread with the reason and then resolve the thread. At .github/workflows/ci.yml, line 359:

<comment>The install matrix omits Python 3.15 even though the project supports it, so this new plain-wheel install check does not cover every supported interpreter. Add 3.15 to the matrix.</comment>

<file context>
@@ -310,6 +310,215 @@ jobs:
+  install-matrix:
+    strategy:
+      matrix:
+        python-version: ["3.10", "3.11", "3.12", "3.13", "3.14"]
+    if: |
+      always() && !cancelled() &&
</file context>

Suggested change

	python-version: ["3.10", "3.11", "3.12", "3.13", "3.14"]
	python-version: ["3.10", "3.11", "3.12", "3.13", "3.14", "3.15"]

[cubic-dev-ai]

P2: Pickling and unpickling these three exceptions fails with a TypeError because their constructors take more than the message. Add a __reduce__ that returns (_rebuild_error, (type(self), self.args, self.__dict__)), as FieldNotLoadedError already does, to RateLimitError, VersionNotSupportedError, and ObjectValidationError. Round-trip coverage in tests/unit/sdk/test_exceptions.py does not include them.

Prompt for AI agents

Check if this issue is valid — if so, understand the root cause and fix it. When an issue isn't valid or won't be fixed in this PR, reply in its thread with the reason and then resolve the thread. At infrahub_sdk/exceptions/base.py, line 209:

<comment>Pickling and unpickling these three exceptions fails with a TypeError because their constructors take more than the message. Add a `__reduce__` that returns `(_rebuild_error, (type(self), self.args, self.__dict__))`, as `FieldNotLoadedError` already does, to `RateLimitError`, `VersionNotSupportedError`, and `ObjectValidationError`. Round-trip coverage in tests/unit/sdk/test_exceptions.py does not include them.</comment>

<file context>
@@ -0,0 +1,522 @@
+        return (self.__class__, (self.failures,))
+
+
+class VersionNotSupportedError(Error):
+    """Raised when a feature is used against an Infrahub server version that does not support it."""
+
</file context>

[cubic-dev-ai]

P2: When OTEL_PROPAGATORS is invalid, this import fails on every request, because the failed module is not cached and the ValueError path only caches the warning. Cache the failure result too, so the import is not re-attempted per request.

Prompt for AI agents

Check if this issue is valid — if so, understand the root cause and fix it. When an issue isn't valid or won't be fixed in this PR, reply in its thread with the reason and then resolve the thread. At infrahub_sdk/trace_context.py, line 19:

<comment>When OTEL_PROPAGATORS is invalid, this import fails on every request, because the failed module is not cached and the `ValueError` path only caches the warning. Cache the failure result too, so the import is not re-attempted per request.</comment>

<file context>
@@ -0,0 +1,54 @@
+        return {}
+
+    try:
+        from opentelemetry import propagate  # noqa: PLC0415
+    except ImportError:
+        return {}
</file context>

[cubic-dev-ai]

P3: This claims every fallback is logged at debug level. On the authentication path, a response with a code that has no class raises AuthenticationError without a log line, because the logging only runs when no code was resolved. Narrow the claim to the cases that actually log.

Prompt for AI agents

Check if this issue is valid — if so, understand the root cause and fix it. When an issue isn't valid or won't be fixed in this PR, reply in its thread with the reason and then resolve the thread. At docs/docs/python-sdk/topics/error_handling.mdx, line 196:

<comment>This claims every fallback is logged at debug level. On the authentication path, a response with a code that has no class raises `AuthenticationError` without a log line, because the logging only runs when no code was resolved. Narrow the claim to the cases that actually log.</comment>

<file context>
@@ -0,0 +1,212 @@
+| A server predating the catalogue, or an error with no `extensions` | `exc.code` is `None`, and the message is the one that version of the SDK has always produced |
+| A payload that does not match what the catalogue declares | The generic class for the transport, with the code still readable |
+
+Every fallback is logged at debug level on the `infrahub_sdk` logger, naming the code where the
+response carried one, so an SDK meeting a newer server is diagnosable without a debugger.
+
</file context>

[cubic-dev-ai]

P3: python_sdk is a git submodule, so Infrahub pins a commit rather than tracking an SDK branch; this wording can make readers expect branch-following behavior. Say that the submodule points to a commit from the matching SDK branch.

Prompt for AI agents

Check if this issue is valid — if so, understand the root cause and fix it. When an issue isn't valid or won't be fixed in this PR, reply in its thread with the reason and then resolve the thread. At AGENTS.md, line 65:

<comment>`python_sdk` is a git submodule, so Infrahub pins a commit rather than tracking an SDK branch; this wording can make readers expect branch-following behavior. Say that the submodule points to a commit from the matching SDK branch.</comment>

<file context>
@@ -53,6 +53,22 @@ infrahub_sdk/
+| `infrahub-release-x.y` | The SDK that ships with Infrahub `x.y`; created from `infrahub-develop` when Infrahub cuts `release-x.y` |
+| `infrahub-develop` | SDK changes that need the next Infrahub version |
+
+The Infrahub repository pins `python_sdk` to the branch that matches its own: `stable` to `stable`,
+`release-x.y` to `infrahub-release-x.y`, `develop` to `infrahub-develop`.
+
</file context>

Suggested change

	The Infrahub repository pins `python_sdk` to the branch that matches its own: `stable` to `stable`,
	`release-x.y` to `infrahub-release-x.y`, `develop` to `infrahub-develop`.
	The Infrahub repository's `python_sdk` submodule points to a commit from the SDK branch that matches its own: `stable` to `stable`,
	`release-x.y` to `infrahub-release-x.y`, `develop` to `infrahub-develop`.

[cubic-dev-ai]

P3: This release note claims to list all new catalogue exception classes but omits the exported WorkerTimeoutError. Add WorkerTimeoutError to the list so users can discover the new WORKER_TIMEOUT exception.

Prompt for AI agents

Check if this issue is valid — if so, understand the root cause and fix it. When an issue isn't valid or won't be fixed in this PR, reply in its thread with the reason and then resolve the thread. At CHANGELOG.md, line 54:

<comment>This release note claims to list all new catalogue exception classes but omits the exported `WorkerTimeoutError`. Add `WorkerTimeoutError` to the list so users can discover the new `WORKER_TIMEOUT` exception.</comment>

<file context>
@@ -11,6 +11,192 @@ This project uses [*towncrier*](https://towncrier.readthedocs.io/) and the chang
+      print("some other failure:", exc.code)
+  ```
+
+  The new classes are `AttributeConstraintViolationError`, `AttributeInvalidTypeError`, `AttributeRequiredError`, `BranchAlreadyMergedError`, `BranchNeedsRebaseError`, `MergeInProgressError`, `MergeRecoveryRequiredError`, `UndefinedError`, and `UniquenessViolationError`. Each is typed exactly as the catalogue declares the payload, so a required field is never optional and needs no guard. They all descend from `GraphQLError`, so no `except` clause stops catching what it catches today. `NODE_NOT_FOUND`, `BRANCH_NOT_FOUND` and `SCHEMA_NOT_FOUND` reach the classes the SDK already had for them, listed under Changed.
+
+  `exc.http_status` reads the status the envelope declares. Where the envelope omits one, a generated class supplies the status the catalogue gives its code; the three lookup-miss classes declare none, since they are also raised with no server behind them, and stay `None` there.
</file context>

Suggested change

The new classes are `AttributeConstraintViolationError`, `AttributeInvalidTypeError`, `AttributeRequiredError`, `BranchAlreadyMergedError`, `BranchNeedsRebaseError`, `MergeInProgressError`, `MergeRecoveryRequiredError`, `UndefinedError`, and `UniquenessViolationError`. Each is typed exactly as the catalogue declares the payload, so a required field is never optional and needs no guard. They all descend from `GraphQLError`, so no `except` clause stops catching what it catches today. `NODE_NOT_FOUND`, `BRANCH_NOT_FOUND` and `SCHEMA_NOT_FOUND` reach the classes the SDK already had for them, listed under Changed.

The new classes are `AttributeConstraintViolationError`, `AttributeInvalidTypeError`, `AttributeRequiredError`, `BranchAlreadyMergedError`, `BranchNeedsRebaseError`, `MergeInProgressError`, `MergeRecoveryRequiredError`, `UndefinedError`, `UniquenessViolationError`, and `WorkerTimeoutError`. Each is typed exactly as the catalogue declares the payload, so a required field is never optional and needs no guard. They all descend from `GraphQLError`, so no `except` clause stops catching what it catches today. `NODE_NOT_FOUND`, `BRANCH_NOT_FOUND` and `SCHEMA_NOT_FOUND` reach the classes the SDK already had for them, listed under Changed.

[cubic-dev-ai]

P3: These new troubleshooting commands cross the Infrahub server boundary, but coverage only verifies mocked HTTP responses. Add a functional or integration test against a compatible Infrahub server to verify the endpoint contract and response handling.

(Based on your team's feedback about requiring real-server coverage for server-bound SDK and CLI behavior.) [ID: 8542b3a4-57c4-4f1c-abe8-ef576231d2c1]

Prompt for AI agents

Check if this issue is valid — if so, understand the root cause and fix it. When an issue isn't valid or won't be fixed in this PR, reply in its thread with the reason and then resolve the thread. At infrahub_sdk/ctl/troubleshooting.py, line 74:

<comment>These new troubleshooting commands cross the Infrahub server boundary, but coverage only verifies mocked HTTP responses. Add a functional or integration test against a compatible Infrahub server to verify the endpoint contract and response handling.

(Based on your team's feedback about requiring real-server coverage for server-bound SDK and CLI behavior.) [ID: 8542b3a4-57c4-4f1c-abe8-ef576231d2c1]</comment>

<file context>
@@ -0,0 +1,102 @@
+async def get_trace_sample_ratio(_: str = CONFIG_PARAM) -> None:
+    """Show the trace sampling ratio of the API server that answers."""
+    client = initialize_client()
+    response = await client._get(url=f"{client.address}{TRACE_SAMPLE_RATIO_PATH}")
+    _print_state(response)
+
</file context>

[cubic-dev-ai]

P3: This SDK reference exposes the internal peers_are_inline_input predicate as a public API. Remove this page, or document the supported RelationshipManager behavior instead; keep internal implementation helpers out of the SDK reference.

Prompt for AI agents

Check if this issue is valid — if so, understand the root cause and fix it. When an issue isn't valid or won't be fixed in this PR, reply in its thread with the reason and then resolve the thread. At docs/docs/python-sdk/sdk_ref/infrahub_sdk/node/inline_peers.mdx, line 2:

<comment>This SDK reference exposes the internal `peers_are_inline_input` predicate as a public API. Remove this page, or document the supported `RelationshipManager` behavior instead; keep internal implementation helpers out of the SDK reference.</comment>

<file context>
@@ -0,0 +1,18 @@
+---
+title: inline_peers
+sidebarTitle: inline_peers
+---
</file context>