Context/example.context at main · Modular-Intellegence-Systems/Context · GitHub

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
@CONTEXT/1.2 profile=human canon=CTX-CANON/3
[ns]
ctx=ctx:
doc=http://spec.example/context#
eval=eval:
exp=exp:
ops=ops:

[meta]
doc=01k8nye3njp97qp90hx7asg66k
author=core-context-team
date=2025-10-28T18:32:00Z
schema=CONTEXT
version=1.2
units=ctx.tokens
lang=en
policy.cf=Confidence derived from codex-exec AB runs (success ratio).
policy.ttl=Review capsules at TTL or on new experiment cycle.

[cap i_ab_result]
t=ctx.I
p=9
cf=0.870
ts=2025-10-28T18:31:00Z
ttl=P120D
src=exp:ab-ctx12
lang=en
tags=analysis,comparison,token-budget
d=Variant beta saved 23% tokens and 11% latency against baseline after applying CONTEXT/1.2 capsules.

[cap k_ctx_graph_schema]
t=ctx.K
p=8
cf=0.930
ts=2025-10-28T18:20:00Z
ttl=P365D
src=spec:ctx-graph
lang=en
tags=context,graph-model,semantics
d=CONTEXT/1.2 namespaces provide deterministic graph edges for capsule relations.

[cap l_spec_anchor]
t=ctx.L
p=5
cf=1.000
ts=2025-10-28T18:18:00Z
ttl=P365D
src=spec:ctx12
lang=en
tags=reference,specification
d=https://intranet.example/specs/context_1_2.html

[cap n_noise_packet]
t=ctx.N
p=2
cf=0.200
ts=2025-10-28T18:10:00Z
ttl=P30D
src=eval:raw-log-771
lang=en
tags=raw,telemetry
d=Latency samples: [125, 119, 421, 118, 122] ms (contains outliers).

[cap s_ingest_pipeline]
t=ctx.S
p=7
cf=0.900
ts=2025-10-28T18:24:00Z
ttl=P180D
src=ops:routing
lang=en
tags=ingestion,pipeline,rag
note=Pipeline assumes canonical CONTEXT feed profile.
d@text/markdown<<PIPE
**Ingestion Steps**:
1. Validate namespace prefixes via ctx_lint.
2. Filter capsules with cf>=0.6 and exclude ctx.N by default.
3. Emit feed profile sorted by priority and ts.
4. Persist per-run digest for audit trail.
PIPE

[cap t_execute_variant]
t=ctx.T
p=8
cf=0.880
ts=2025-10-28T18:29:30Z
ttl=P30D
src=ops:execution
lang=en
tags=trace,workflow
in=t_plan_variant,k_ctx_graph_schema,@01jdzr7v95kq7b9n91hza7ap8d#i_token_saving
out=i_ab_result
op=ctx.execute
cost.kind=ctx.ms
cost.val=842.000
d=Ran variant beta against baseline using namespace-aware retrieval.

[cap t_plan_variant]
t=ctx.T
p=8
cf=0.910
ts=2025-10-28T18:26:00Z
ttl=P30D
src=ops:planning
lang=en
tags=trace,workflow
note=Prepare evaluation sequence before execution.
in=L"objective\=compare variant-beta",s_ingest_pipeline
out=t_execute_variant
op=ctx.plan
cost.kind=ctx.tokens
cost.val=96
d=Planned experiment harness using CONTEXT/1.2 feed generation.

[trace ch_variant_eval]
goal=i_ab_result
head=t_plan_variant
halt=t_execute_variant
status=ctx.ok
ts=2025-10-28T18:29:55Z
tags=analysis,reasoning

[rel]
r=i_ab_result|ctx.cites|l_spec_anchor
r=i_ab_result|ctx.supports|@01jdzr7v95kq7b9n91hza7ap8d#i_token_saving
r=k_ctx_graph_schema|ctx.supports|t_execute_variant|w=0.8|ts=2025-10-28T18:29:00Z
r=n_noise_packet|ctx.clarifies|i_ab_result|w=0.2
r=s_ingest_pipeline|ctx.applied_by|t_plan_variant
r=t_execute_variant|ctx.derived_from|k_ctx_graph_schema
r=t_plan_variant|ctx.applied_by|t_execute_variant

[footer]
digest=sha256
digest-base16=dcf4120b720207b8d54f308abfe0a799f3f50ad2f474ea99059066f647f5f64d