Files
wehub-resource-sync 4b6817381b
CI (OpenClaw E2E) / openclaw test (push) Has been cancelled
CI / coverage-report (push) Has been cancelled
CI / test-kubernetes (push) Has been cancelled
CI / should-run-thorough (push) Has been cancelled
CI / test-thorough (cloudwatch-demo) (push) Has been cancelled
CI / test-thorough (flink-ecs) (push) Has been cancelled
CI / test-thorough (upstream-lambda) (push) Has been cancelled
CI / test-thorough (prefect-ecs-fargate) (push) Has been cancelled
Release / build-binaries (zip, opensre.exe, onefile, windows-latest, windows-x64) (push) Has been cancelled
Benchmark image — build + push to ECR (any adapter) / build + push (push) Has been cancelled
CI / quality (ubuntu-latest) (push) Has been cancelled
CI / test (tools-runtime) (push) Has been cancelled
CI / test (e2e-general) (push) Has been cancelled
CI / test (cli-runtime) (push) Has been cancelled
CI / test (e2e-provider-and-openclaw) (push) Has been cancelled
CI / test (integrations-and-misc) (push) Has been cancelled
Release / verify (push) Has been cancelled
Release / build-python-dist (push) Has been cancelled
Release / build-binaries (tar.gz, opensre, onedir, macos-15-intel, darwin-x64) (push) Has been cancelled
Release / build-binaries (tar.gz, opensre, onedir, macos-latest, darwin-arm64) (push) Has been cancelled
Release / build-binaries (tar.gz, opensre, onedir, ubuntu-22.04, linux-x64) (push) Has been cancelled
Release / publish-release (push) Has been cancelled
Release / publish-main-release (push) Has been cancelled
Interactive Shell Live (PR + post-merge) / turn-checks (no-LLM) (push) Has been cancelled
CodeQL / Analyze (python) (push) Has been cancelled
Interactive Shell Live (PR + post-merge) / turn-live shard ${{ matrix.shard_index }} (push) Has been cancelled
Release / prepare (push) Has been cancelled
Release / build-binaries (tar.gz, opensre, onedir, ubuntu-22.04-arm, linux-arm64) (push) Has been cancelled
Synthetic Deterministic Tests / Synthetic offline (deterministic) (push) Has been cancelled
chore: import upstream snapshot with attribution
2026-07-13 13:10:45 +08:00

280 lines
10 KiB
Python

# ======== from tools/groundcover_logs_tool/ ========
"""groundcover logs query tool (gcQL over query_logs)."""
from __future__ import annotations
from typing import Any
from core.tool_framework.tool_decorator import tool
from integrations.groundcover.availability import groundcover_available_or_backend
from integrations.groundcover.client import GroundcoverClient
from integrations.groundcover.helpers import (
DEFAULT_LOGS_QUERY,
GCQL_GUIDANCE,
base_extract_params,
run_signal_query,
)
_LOGS_SOURCE = "groundcover_logs"
_LOGS_MCP_TOOL = "query_logs"
_LOGS_QUERY_DESCRIPTION = (
"gcQL query. Lead with the filter directly (not a '| filter' pipe) and include "
"'| limit N'. Project raw rows with '| fields ...' rather than a bare select-all. "
"Examples: 'level:error | fields _time, workload, instance, content | limit 50'; "
"'workload:checkout level:error | fields _time, instance, content | limit 50'; "
"'* | stats by (workload) count() if (level:error) as errors | sort by (errors desc) "
"| limit 20'."
)
def _logs_is_available(sources: dict[str, dict]) -> bool:
"""Available when groundcover credentials are present or a fixture backend is injected."""
return groundcover_available_or_backend(sources)
def _logs_extract_params(sources: dict[str, dict]) -> dict[str, Any]:
"""Inject a pre-built client + seed query, never raw credentials."""
return base_extract_params(sources.get("groundcover", {}), default_query=DEFAULT_LOGS_QUERY)
@tool(
name="query_groundcover_logs",
display_name="groundcover logs",
source="groundcover",
tags=("logs", "observability"),
description=(
"Search groundcover logs with gcQL. Use for application errors, exceptions, and service "
"log events. " + GCQL_GUIDANCE + " Discover fields with '* | field_names' or by calling "
"get_groundcover_query_reference."
),
use_cases=[
"Finding error/exception logs for a workload or namespace",
"Correlating log spikes with a groundcover monitor issue",
"Counting errors per workload with a single stats query over a narrow window",
],
requires=[],
input_schema={
"type": "object",
"properties": {
"query": {"type": "string", "description": _LOGS_QUERY_DESCRIPTION},
"start": {"type": "string", "description": "RFC3339 start time (optional)"},
"end": {"type": "string", "description": "RFC3339 end time (optional)"},
"period": {
"type": "string",
"description": "ISO-8601 duration window, e.g. PT1H (default).",
"default": "PT1H",
},
},
"required": ["query"],
"additionalProperties": False,
},
is_available=_logs_is_available,
extract_params=_logs_extract_params,
)
def query_groundcover_logs(
query: str = "",
start: str = "",
end: str = "",
period: str = "",
_groundcover_client: GroundcoverClient | None = None,
groundcover_backend: Any = None,
**_kwargs: Any,
) -> dict[str, Any]:
"""Search groundcover logs with gcQL and return the normalized OpenSRE envelope.
Credentials never travel through the model-facing arguments: ``extract_params``
binds a pre-built :class:`GroundcoverClient` into ``_groundcover_client`` (and an
optional synthetic ``groundcover_backend``), both stripped from seed input by the
redactor. An empty query yields a cheap guidance envelope with no MCP round trip.
"""
return run_signal_query(
source=_LOGS_SOURCE,
mcp_tool=_LOGS_MCP_TOOL,
client=_groundcover_client,
query=query,
start=start,
end=end,
period=period,
backend=groundcover_backend,
)
# ======== from tools/groundcover_query_reference_tool/ ========
"""groundcover query-language (gcQL) reference tool."""
from typing import cast
from core.tool_framework.tool_decorator import tool
_QUERY_REF_SOURCE = "groundcover_query_reference"
def _query_ref_is_available(sources: dict[str, dict]) -> bool:
"""Available when groundcover credentials are present or a fixture backend is injected."""
return groundcover_available_or_backend(sources)
def _query_ref_extract_params(sources: dict[str, dict]) -> dict[str, Any]:
"""Inject a pre-built client (no time window), never raw credentials."""
return base_extract_params(sources.get("groundcover", {}), include_period=False)
@tool(
name="get_groundcover_query_reference",
display_name="groundcover query reference",
source="groundcover",
tags=("observability", "reference"),
surfaces=("investigation", "chat"),
description=(
"Get the groundcover Query Language (gcQL) reference: operators, functions, pipes, and "
"query patterns. Call this ONCE before writing gcQL for any query_groundcover_* tool. "
"Reading it first prevents malformed and overly expensive queries."
),
use_cases=[
"Before composing any non-trivial gcQL query for groundcover logs/traces/metrics/apm",
"When unsure about stats/sort/filter syntax or pipe operators",
"To recall the performance and time-window guidance for efficient queries",
],
requires=[],
input_schema={"type": "object", "properties": {}, "additionalProperties": False},
is_available=_query_ref_is_available,
extract_params=_query_ref_extract_params,
)
def get_groundcover_query_reference(
_groundcover_client: GroundcoverClient | None = None,
groundcover_backend: Any = None,
**_kwargs: Any,
) -> dict[str, Any]:
"""Return the cached gcQL reference skill content."""
if groundcover_backend is not None and hasattr(groundcover_backend, "get_query_reference"):
return cast("dict[str, Any]", groundcover_backend.get_query_reference())
if _groundcover_client is None:
return {
"source": _QUERY_REF_SOURCE,
"available": False,
"reference": "",
"error": "groundcover integration not configured",
}
result = _groundcover_client.get_query_reference()
if not result.get("success"):
return {
"source": _QUERY_REF_SOURCE,
"available": False,
"reference": "",
"error": result.get("error", "could not fetch gcQL reference"),
}
return {
"source": _QUERY_REF_SOURCE,
"available": True,
"reference": result.get("reference", ""),
"cached": result.get("cached", False),
"error": None,
}
# ======== from tools/groundcover_traces_tool/ ========
"""groundcover traces query tool (gcQL over query_traces)."""
from core.tool_framework.tool_decorator import tool
from integrations.groundcover.helpers import (
DEFAULT_TRACES_QUERY,
GCQL_GUIDANCE,
)
_TRACES_SOURCE = "groundcover_traces"
_TRACES_MCP_TOOL = "query_traces"
_TRACES_QUERY_DESCRIPTION = (
"gcQL query. Lead with the filter directly (not a '| filter' pipe) and include "
"'| limit N'. Project raw spans with '| fields ...' (a bare select-all '| limit N' is "
"rejected for traces); otherwise aggregate with '| stats ...'. Examples: "
"'workload:checkout duration_seconds>0.5 | fields _time, span_name, duration_seconds "
"| sort by (duration_seconds desc) | limit 50'; "
"'status_code>=500 | stats by (workload) count() as errors | sort by (errors desc) "
"| limit 20'; "
"'span_type:mysql status:error | fields _time, span_name, status_code | limit 50'."
)
def _traces_is_available(sources: dict[str, dict]) -> bool:
"""Available when groundcover credentials are present or a fixture backend is injected."""
return groundcover_available_or_backend(sources)
def _traces_extract_params(sources: dict[str, dict]) -> dict[str, Any]:
"""Inject a pre-built client + seed query, never raw credentials."""
return base_extract_params(sources.get("groundcover", {}), default_query=DEFAULT_TRACES_QUERY)
@tool(
name="query_groundcover_traces",
display_name="groundcover traces",
source="groundcover",
tags=("traces", "observability"),
description=(
"Query groundcover traces/spans with gcQL. Use to find slow spans, failing spans, and "
"request correlations across services. " + GCQL_GUIDANCE + " Discover fields with "
"'* | field_names'. For traces, free text needs '*:*term*' or 'field:*term*' (no bare "
"keywords). Error filtering by span type: HTTP spans use 'status_code>=500' (or "
"'status_code>399'); databases/gRPC/messaging and any span type use 'status:error' "
"(universal). Never use 'status_code>399' on non-HTTP spans — their codes differ. Key "
"fields: span_name (endpoint), workload (caller), server (callee), duration_seconds, status."
),
use_cases=[
"Finding the slowest spans for a workload (sort by duration_seconds desc)",
"Locating 5xx/erroring spans for a service or endpoint",
"Aggregating error rate and p95 latency per workload with one stats query",
],
requires=[],
input_schema={
"type": "object",
"properties": {
"query": {"type": "string", "description": _TRACES_QUERY_DESCRIPTION},
"start": {"type": "string", "description": "RFC3339 start time (optional)"},
"end": {"type": "string", "description": "RFC3339 end time (optional)"},
"period": {
"type": "string",
"description": "ISO-8601 duration window, e.g. PT1H (default).",
"default": "PT1H",
},
},
"required": ["query"],
"additionalProperties": False,
},
is_available=_traces_is_available,
extract_params=_traces_extract_params,
)
def query_groundcover_traces(
query: str = "",
start: str = "",
end: str = "",
period: str = "",
_groundcover_client: GroundcoverClient | None = None,
groundcover_backend: Any = None,
**_kwargs: Any,
) -> dict[str, Any]:
"""Query groundcover traces/spans with gcQL and return the normalized OpenSRE envelope.
Credentials never travel through the model-facing arguments: ``extract_params``
binds a pre-built :class:`GroundcoverClient` into ``_groundcover_client`` (and an
optional synthetic ``groundcover_backend``), both stripped from seed input by the
redactor. An empty query yields a cheap guidance envelope with no MCP round trip.
"""
return run_signal_query(
source=_TRACES_SOURCE,
mcp_tool=_TRACES_MCP_TOOL,
client=_groundcover_client,
query=query,
start=start,
end=end,
period=period,
backend=groundcover_backend,
)