Hugging Face's logo

Spaces:
eastbrick
/
releaseops-env
Sleeping

App Files Community
releaseops-env
File size: 5,820 Bytes
408d02c
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
"""ReleaseOps-Env: Typed models for production change review environment."""

from typing import Any, Literal, Optional
from pydantic import BaseModel, Field
from openenv.core.env_server.types import Action, Observation, State


# ──────────────────────────────────────────────
# ACTION MODEL
# ──────────────────────────────────────────────


class ReleaseAction(Action):
    """
    Single action type with discriminator routing.

    action_type determines which fields are used:
    - "inspect_change": section
    - "inspect_services": service
    - "inspect_dependencies": service
    - "search_incidents": keywords
    - "check_policy": (no extra fields)
    - "query_telemetry": metric, service, window
    - "request_artifact": artifact_type
    - "control_rollout": decision
    - "submit_decision": final_decision, reason_codes
    """

    action_type: Literal[
        "inspect_change",
        "inspect_services",
        "inspect_dependencies",
        "search_incidents",
        "check_policy",
        "query_telemetry",
        "request_artifact",
        "control_rollout",
        "submit_decision",
    ]

    # inspect_change
    section: Optional[Literal["diff", "tests", "approvals", "files_changed"]] = None

    # inspect_services, inspect_dependencies, query_telemetry
    service: Optional[str] = None

    # search_incidents
    keywords: Optional[list[str]] = None

    # query_telemetry
    metric: Optional[
        Literal["p50", "p95", "p99", "error_rate", "queue_depth", "cpu", "rps"]
    ] = None
    window: Optional[Literal["5m", "15m", "1h"]] = None

    # request_artifact
    artifact_type: Optional[
        Literal[
            "load_test",
            "rollback_plan",
            "approval",
            "runbook",
            "security_review",
            "compliance_check",
        ]
    ] = None

    # control_rollout
    decision: Optional[Literal["start_canary", "pause", "promote", "rollback"]] = None

    # submit_decision
    final_decision: Optional[
        Literal["approve", "request_changes", "block", "escalate"]
    ] = None
    reason_codes: Optional[list[str]] = None


# ──────────────────────────────────────────────
# RISK SIGNAL (sub-model for observations)
# ──────────────────────────────────────────────


class RiskSignal(BaseModel):
    """A discovered risk signal surfaced through investigation."""

    signal_id: str
    category: Literal[
        "latency",
        "error_rate",
        "dependency_health",
        "test_coverage",
        "blast_radius",
        "policy_violation",
        "missing_approval",
        "incident_similarity",
        "compliance",
        "resource_saturation",
    ]
    severity: Literal["low", "medium", "high", "critical"]
    summary: str
    numeric_value: Optional[float] = None
    threshold: Optional[float] = None


# ──────────────────────────────────────────────
# TOOL RESULT (sub-model for observations)
# ──────────────────────────────────────────────


class ToolResult(BaseModel):
    """Result from the last tool/action invocation."""

    tool_name: str
    success: bool
    content: str
    structured_payload: dict[str, Any] = Field(default_factory=dict)


# ──────────────────────────────────────────────
# OBSERVATION MODEL
# ──────────────────────────────────────────────


class ReleaseObservation(Observation):
    """
    What the agent sees after each step.

    Starts sparse (just change_summary) and fills in as agent investigates.
    """

    # Context
    task_id: str = ""
    change_summary: str = ""

    # Discovery state
    known_risk_signals: list[RiskSignal] = Field(default_factory=list)
    last_tool_result: Optional[ToolResult] = None

    # Navigation
    allowed_actions: list[str] = Field(default_factory=list)
    rollout_phase: str = "precheck"

    # Budget
    time_remaining: int = 12
    cumulative_reward: float = 0.0

    # Terminal
    final_score: Optional[float] = None
    grader_breakdown: Optional[dict[str, float]] = None


# ──────────────────────────────────────────────
# STATE MODEL
# ──────────────────────────────────────────────


class ReleaseState(State):
    """
    Full environment state (server-side).

    Contains both public trajectory info and hidden simulator truth.
    The agent never sees hidden_* fields — only the Observation is sent.
    """

    # Public tracking
    task_id: str = ""
    rollout_phase: Literal[
        "precheck", "canary", "paused", "promoted", "rolled_back", "terminal"
    ] = "precheck"
    steps_in_phase: int = 0
    evidence_gathered: list[str] = Field(default_factory=list)
    risk_signals_found: list[str] = Field(default_factory=list)
    artifacts_requested: list[str] = Field(default_factory=list)
    actions_taken: list[str] = Field(default_factory=list)
    cumulative_reward: float = 0.0
    terminal: bool = False
    final_score: Optional[float] = None