serbero 0.1.1

Nostr-native dispute coordination daemon for the Mostro ecosystem
Documentation 
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

# Serbero sample configuration.
#
# Usage:
# 1. Copy this file to `config.toml`
# 2. Replace the placeholder hex keys / relay URLs / solver pubkeys
# 3. Export secrets through environment variables instead of committing them
#
# The daemon loads `./config.toml` by default, or the path set in
# `SERBERO_CONFIG`.

[serbero]
# Hex-encoded Nostr private key for the Serbero daemon identity.
# Production deployments should prefer `SERBERO_PRIVATE_KEY`.
private_key = "0123456789abcdef0123456789abcdef0123456789abcdef0123456789abcdef"

# SQLite database path. Can be overridden with `SERBERO_DB_PATH`.
db_path = "serbero.db"

# Tracing log level / filter. Can be overridden with `SERBERO_LOG`.
log_level = "info"

[mostro]
# Hex-encoded public key of the Mostro instance to monitor.
pubkey = "abcdef0123456789abcdef0123456789abcdef0123456789abcdef0123456789"

[[relays]]
url = "wss://relay.example.com"

# Add as many relays as needed.
[[relays]]
url = "wss://relay.backup.example.com"

# Solvers notified for dispute events.
# `permission` is parsed today and becomes more relevant in later phases.
[[solvers]]
pubkey = "1111111111111111111111111111111111111111111111111111111111111111"
permission = "read"

[[solvers]]
pubkey = "2222222222222222222222222222222222222222222222222222222222222222"
permission = "write"

[timeouts]
# Re-notify disputes left in `notified` state for this long.
renotification_seconds = 300

# How often to scan for unattended disputes.
renotification_check_interval_seconds = 60

# ---------------------------------------------------------------------------
# Phase 3: guided mediation
# Omit or keep disabled to run only Phase 1 / 2 behavior.
# ---------------------------------------------------------------------------

[mediation]
enabled = false

# Maximum mediation rounds before `round_limit` escalation.
max_rounds = 2

# Escalate if a party does not respond within this many seconds.
# Set to 0 to disable the timeout sweep.
party_response_timeout_seconds = 1800

# Solver-auth revalidation loop.
solver_auth_retry_initial_seconds = 60
solver_auth_retry_max_interval_seconds = 3600
solver_auth_retry_max_total_seconds = 86400
solver_auth_retry_max_attempts = 24

[reasoning]
enabled = false

# Phase 3 currently ships the OpenAI-compatible adapter.
provider = "openai"

# Any model accepted by the configured OpenAI-compatible endpoint.
model = "gpt-5"

# Hosted OpenAI example. Replace with any OpenAI-compatible endpoint if needed.
api_base = "https://api.openai.com/v1"

# Name of the environment variable holding the reasoning API credential.
# Example:
#   export SERBERO_REASONING_API_KEY="your-real-key"
api_key_env = "SERBERO_REASONING_API_KEY"

# Per-request timeout and bounded retry budget.
request_timeout_seconds = 30
followup_retry_count = 1

[prompts]
# Default prompt bundle paths shipped in this repository.
system_instructions_path = "./prompts/phase3-system.md"
classification_policy_path = "./prompts/phase3-classification.md"
escalation_policy_path = "./prompts/phase3-escalation-policy.md"
mediation_style_path = "./prompts/phase3-mediation-style.md"
message_templates_path = "./prompts/phase3-message-templates.md"

[chat]
# Inbound Mostro chat polling interval used by the mediation engine.
inbound_fetch_interval_seconds = 10

# --- Phase 4: escalation execution surface (FR-215) ---
# Consumes Phase 3's handoff_prepared audit events and dispatches
# a structured DM to write-permission solvers. See
# specs/004-escalation-execution/contracts/config.md for the full
# shape.
[escalation]
# Feature flag. `true` spawns the Phase 4 dispatcher task; `false`
# keeps Phase 4 entirely inert (no task, no writes, no Phase 1/2/3
# behavior change). Default: false — Phase 4 is opt-in.
enabled = false

# Scan cadence. Positive integer seconds. The SC-201 target (95% of
# handoffs delivered within 60 s) assumes the default 30 here;
# larger values degrade delivery latency linearly.
dispatch_interval_seconds = 30

# When zero solvers with `permission = "write"` are configured:
#   true  → broadcast the handoff DM to every configured solver
#           regardless of permission.
#   false → refuse to broadcast and record an
#           `escalation_dispatch_unroutable` audit event + ERROR log
#           line. The handoff stays pending so a later config
#           change (adding a write solver) re-surfaces it.
# Operators running intentionally read-only deployments must opt in
# explicitly by flipping this to true.
fallback_to_all_solvers = false