Complete Docker ecosystem integration and common LLM runtime flow - #20

Merged

abmmhasan merged 509 commits into

mainfrom

plan/docker-ecosystem-bottom-up

Sep 21, 2026

This pull request is big! We're only showing the most recent 250 commits

Commits on Sep 19, 2026

fix(ai): set local Graphify Ollama API key
abmmhasan
committed
6546fb0
test(cli): assert local Graphify API key
abmmhasan
committed
694f404
test(ai): lock Graphify local API key
abmmhasan
committed
f471f11
docs(ai): explain Graphify local API key
abmmhasan
committed
64218ce
docs(ai): document Graphify API key placeholder
abmmhasan
committed
2f02084
refactor(ai): scope Graphify environment once
abmmhasan
committed
aaf4418
feat(ai): detect AMD CPU iGPU preference
abmmhasan
committed
b3d67bd
feat(ai): persist AMD iGPU preference during setup
abmmhasan
committed
7dadfb0
feat(ai): align qwen3 default and AMD iGPU runtime state
abmmhasan
committed
0f491ef
feat(ai): forward AMD iGPU enablement to Ollama
abmmhasan
committed
f98a8f7
feat(ai): default profile model to qwen3 14b
abmmhasan
committed
91f2e76
test(ai): make fake Ollama advertise qwen3 14b
abmmhasan
committed
111eef6
test(ai): align CLI contract with qwen3 14b
abmmhasan
committed
ad72142
test(ai): align provider contract with qwen3 14b
abmmhasan
committed
c27d944
test(docs): expect qwen3 14b AI default
abmmhasan
committed
23d06f1
test(ai): cover qwen3 default and iGPU forwarding
abmmhasan
committed
4497f24
test(ai): cover AMD CPU iGPU persistence
abmmhasan
committed
8f5a1ad
test(ai): lock iGPU forwarding and runtime persistence
abmmhasan
committed
a323dd3
docs(ai): document qwen3 default and AMD iGPU enablement
abmmhasan
committed
b42c4d5
docs(ai): align local runtime guide with qwen3 and AMD iGPU
abmmhasan
committed
f1949f6
docs(plan): sync qwen3 default and AMD iGPU behavior
abmmhasan
committed
e0e4059
docs(ai): describe qwen3 and AMD iGPU profile settings
abmmhasan
committed
6878f5f
docs(ai): document AMD iGPU provider wiring
abmmhasan
committed
809d690
docs(cli): document AMD iGPU runtime state
abmmhasan
committed
68380a5
test(ai): align catalog default with qwen3 14b
abmmhasan
committed
af15a90
docs(ai): consume renamed llm-ollama image
abmmhasan
committed
233ed33
feat(ai): consume infocyph/llm-ollama
abmmhasan
committed
4b2d88c
chore(ai): report llm-ollama image name
abmmhasan
committed
9be5997
test(ai): expect llm-ollama provider image
abmmhasan
committed
7b1802d
test(ai): align compose image contract with llm-ollama
abmmhasan
committed
e23cad7
test(images): track llm-ollama provider image
abmmhasan
committed
f994dd3
docs(ai): rename provider image to llm-ollama
abmmhasan
committed
1513e90
docs(ai): rename provider image in architecture
abmmhasan
committed
4c270fc
docs(ai): rename provider image in environment guide
abmmhasan
committed
0040db2
docs(plan): rename llm provider repository
abmmhasan
committed
aa495a7
refactor(ai): rename LocalDevStack Ollama runtime contract
abmmhasan
committed
3674cd4
refactor(ai): rename LocalDevStack Ollama runtime contract
abmmhasan
committed
f27ac3f
refactor(ai): rename LocalDevStack Ollama runtime contract
abmmhasan
committed
ab3bc3f
refactor(ai): rename LocalDevStack Ollama runtime contract
abmmhasan
committed
49adb5c
refactor(ai): rename LocalDevStack Ollama runtime contract
abmmhasan
committed
9e1ae9d
refactor(ai): rename LocalDevStack Ollama runtime contract
abmmhasan
committed
5dbf522
refactor(ai): rename LocalDevStack Ollama runtime contract
abmmhasan
committed
089f410
refactor(ai): rename LocalDevStack Ollama runtime contract
abmmhasan
committed
ba44caa
docs(ai): align llm-ollama runtime and domain contract
abmmhasan
committed
a37117c
docs(ai): align llm-ollama runtime and domain contract
abmmhasan
committed
9dd7e2c
docs(ai): align llm-ollama runtime and domain contract
abmmhasan
committed
0e08cdf
docs(ai): align llm-ollama runtime and domain contract
abmmhasan
committed
824963a
docs(ai): align llm-ollama runtime and domain contract
abmmhasan
committed
15f5bfd
docs(ai): align llm-ollama runtime and domain contract
abmmhasan
committed
858810a
docs(ai): align llm-ollama runtime and domain contract
abmmhasan
committed
d17b2d7
docs(ai): align llm-ollama runtime and domain contract
abmmhasan
committed
950fbf9
docs(ai): finish llm-ollama rename
abmmhasan
committed
a9f258e
docs(ai): finish llm-ollama rename
abmmhasan
committed
2353625
docs(plan): rename Ollama provider plan file
abmmhasan
committed
f24fa7a
docs(plan): remove old llm-sm plan filename
abmmhasan
committed
df7cdd5
test(ai): finish llm-ollama rename coverage
abmmhasan
committed
5121a7f
test(ai): finish llm-ollama rename coverage
abmmhasan
committed
5a32daf
feat(ai): publish native Ollama port through Nginx
abmmhasan
committed
353bc48
refactor(ai): remove direct provider host-port override
abmmhasan
committed
f2bb2fc
refactor(ai): route Graphify through Nginx native Ollama endpoint
abmmhasan
committed
a567334
docs(ai): make Nginx the sole native Ollama port owner
abmmhasan
committed
dfa4d64
docs(ai): document fixed Nginx Ollama endpoint
abmmhasan
committed
aa1d06e
docs(ai): remove obsolete host-port settings
abmmhasan
committed
6504ae9
docs(cli): remove llm host-port command
abmmhasan
committed
6454225
refactor(ai): remove host-port setup option
abmmhasan
committed
93e74c1
test(ai): remove obsolete host-port catalog contract
abmmhasan
committed
7590ab4
test(ai): assert Nginx owns native Ollama port
abmmhasan
committed
bc4bca9
test(docs): align native Ollama endpoint contract
abmmhasan
committed
fd875f7
docs(cli): remove obsolete llm host-port help
abmmhasan
committed
feeec2d
test(ai): lock Graphify Nginx endpoint contract
abmmhasan
committed
256fc34
docs(plan): make Nginx sole Ollama host-access owner
abmmhasan
committed
3eeddd0
docs(plan): remove provider host-port controls
abmmhasan
committed
68b5caf
docs(ai): remove retired direct-port workflow
abmmhasan
committed
a013e18
docs(cli): point Graphify at fixed Nginx Ollama endpoint
abmmhasan
committed
b42b754
feat(graphify): preflight Nginx native Ollama endpoint
abmmhasan
committed
cfa02ad
test(ai): lock Nginx-owned native Ollama endpoint
abmmhasan
committed
00ba336
docs(plan): finish LLM-Ollama naming
abmmhasan
committed
9cd9672
test(env): remove retired LLM host-port contract
abmmhasan
committed
59e61a5
docs: rename reserved Ollama domain
abmmhasan
committed
b4a2ebf
docs(ai): describe both Nginx Ollama endpoints
abmmhasan
committed
298a2d4
docs(tls): rename Ollama convenience host
abmmhasan
committed
a3b0079
refactor(ai): align diagnostics with llm-ollama
abmmhasan
committed
f0c5699
refactor(ai): remove host-port setup handling
abmmhasan
committed
f2786d5
refactor(ai): rename open shortcut to llm-ollama
abmmhasan
committed
bb2542d
docs(ai): fix runtime literal block formatting
abmmhasan
committed
10f470d

Commits on Sep 20, 2026

fix(cli): preserve proxied stdin and bundle paths
abmmhasan
committed
c265d0b
fix(profiles): make reselection preserve user state
abmmhasan
committed
886c403
fix(profiles): report and validate effective compose profiles
abmmhasan
committed
3baead7
fix(ai): fail fast when Graphify model is unavailable
abmmhasan
committed
70027b8
test(profiles): cover NONE and configured-value preservation
abmmhasan
committed
01568b4
chore(images): move Elastic defaults to latest
abmmhasan
committed
7d6c3e6
chore(images): use latest Elasticsearch by default
abmmhasan
committed
bf3a696
chore(images): use latest Elastic clients by default
abmmhasan
committed
71a6cde
chore(images): report latest Elastic default
abmmhasan
committed
e1f8762
test(images): require latest Elastic catalog default
abmmhasan
committed
9d233e0
test(images): require latest Elastic compose images
abmmhasan
committed
5d28483
test(images): align image output with latest Elastic
abmmhasan
committed
693b0d9
test(bundle): cover relative archive destinations
abmmhasan
committed
6ca3327
test(cli): lock proxied stdin behavior
abmmhasan
committed
0539766
test(ai): cover Graphify model preflight
abmmhasan
committed
7a3e8be
docs: align profiles Graphify and latest image policy
abmmhasan
committed
c8187b5
docs: align profiles Graphify and latest image policy
abmmhasan
committed
536ce8b
docs(profiles): document replacement and current-value semantics
abmmhasan
committed
a68e566
docs(ai): document Graphify model preflight
abmmhasan
committed
102abb1
docs(plan): adopt latest Elastic moving-tag policy
abmmhasan
committed
ee76822
test(images): verify every moving latest alias
abmmhasan
committed
67a0a80
fix(profiles): align menu labels with NONE and cancel semantics
abmmhasan
committed
1ee59c9
fix(profiles): parse cancel and back selectors
abmmhasan
committed
9b50163
chore(images): use current Elastic 9.5.4
abmmhasan
committed
59ecf3f
chore(images): use current Elasticsearch release
abmmhasan
committed
a2fd7c7
chore(images): align Kibana and Filebeat 9.5.4
abmmhasan
committed
abd9e6a
chore(images): report current Elastic release
abmmhasan
committed
7a5533b
test(images): require current Elastic release
abmmhasan
committed
57d1ade
test(images): align Elastic 9.5.4 contracts
abmmhasan
committed
5671316
test(images): report current Elastic release
abmmhasan
committed
21c8841
test(images): verify current Elastic release references
abmmhasan
committed
f3c0ebb
test(images): use resolvable current Elastic release
abmmhasan
committed
c69b4c1
docs(images): document current Elastic release exception
abmmhasan
committed
ac86cd0
docs(images): align Elastic exception with 9.5.4
abmmhasan
committed
9df3a1c
docs(plan): restore verified Elastic current-release exception
abmmhasan
committed
7e574c8
fix(images): use official Elastic registry for Elasticsearch
abmmhasan
committed
3366431
fix(images): use official Elastic registry for Kibana
abmmhasan
committed
9d0ab7c
test(images): validate official Elastic registry references
abmmhasan
committed
e241750
test(images): report official Elasticsearch reference
abmmhasan
committed
6785d50
test(images): resolve official Elastic registry images
abmmhasan
committed
18a9ee8
fix(images): report official Elastic image references
abmmhasan
committed
06dbf24
test(images): use buildx registry inspection for current references
abmmhasan
committed
c9d7c92
feat(ai): prefer supported XDNA2 NPU runtime
abmmhasan
committed
ca996e6
feat(ai): add mutually selected Ollama and FastFlow services
abmmhasan
committed
f3681c7
feat(ai): add persistent FastFlow model volume
abmmhasan
committed
e96e5de
feat(ai): select one provider per runtime in Compose
abmmhasan
committed
d28d4c5
compat(ai): keep Ollama as bare-compose profile fallback
abmmhasan
committed
a384f4f
feat(ai): dispatch LLM CLI and Graphify through active provider
abmmhasan
committed
8e914e3
fix(ai): make service catalog provider-neutral
abmmhasan
committed
d6c7fa8
fix(ai): report and validate provider-aware setup
abmmhasan
committed
c005e44
fix(ai): route user-facing open command through common llm
abmmhasan
committed
f849b02
fix(ai): make diagnostics provider-aware
abmmhasan
committed
a2ac71f
test(ai): make catalog contract provider-neutral
abmmhasan
committed
528cae8
test(ai): expect common Graphify llm route
abmmhasan
committed
e509140
test(ai): make QoL output follow common llm identity
abmmhasan
committed
602e2c6
test(images): include published FastFlow provider
abmmhasan
committed
1c3f830
test(ai): lock NPU provider and model mappings
abmmhasan
committed
c428a6e
test(ai): validate mutually exclusive provider Compose states
abmmhasan
committed
b2f2a07
fix(ai): validate Graphify model through common OpenAI catalog
abmmhasan
committed
445153b
test(ai): replace Ollama-only contract with common llm contract
abmmhasan
committed
f3566f1
ci(ai): rename provider-neutral LLM contract
abmmhasan
committed
81d76c2
docs(ai): describe common mutually exclusive provider flow
abmmhasan
committed
3406da5
docs(ai): align README with mutually exclusive providers
abmmhasan
committed
bc4c14a
docs(ai): rewrite guide for common mutually exclusive providers
abmmhasan
committed
762a0bd
docs(ai): align profile and environment reference
abmmhasan
committed
4f2afed
docs(ai): align CLI reference with common provider flow
abmmhasan
committed
4bf3eb0
test(docs): lock common LLM architecture documentation
abmmhasan
committed
54eb1b4
docs(ai): align architecture with selected common provider
abmmhasan
committed
a35bc97
docs(ai): document both provider model stores
abmmhasan
committed
1a11d63
docs(ai): align README image policy
abmmhasan
committed
c82890d
docs(plan): finalize mutually exclusive LLM provider architecture
abmmhasan
committed
0a5374c
docs(ai): state provider trust boundary explicitly
abmmhasan
committed
dca95c2
test(ai): name provider-neutral trust boundary
abmmhasan
committed
5540b5e
test(docs): align final AI Compose wording
abmmhasan
committed
ac5b911
feat(ai): resolve common llm service alias operationally
abmmhasan
committed
646839d
test(ai): cover common llm operational alias
abmmhasan
committed
cf7d94d
docs(ai): document common llm operational alias
abmmhasan
committed
558644e
docs(ai): prefer common llm alias for operations
abmmhasan
committed
3a387a3
test(docs): lock common llm operational alias
abmmhasan
committed
008ef9c
docs(plan): remove stale Ollama-only completion wording
abmmhasan
committed
bbd5a8a
test(docs): lock final provider architecture plan
abmmhasan
committed
1971985
fix(ai): use Graphify OpenAI backend for FastFlow
abmmhasan
committed
0ca122e
test(ai): cover provider-aware Graphify backends
abmmhasan
committed
f437c93
test(ai): lock Graphify provider backend mapping
abmmhasan
committed
5680d63
docs(ai): describe provider-aware Graphify backend
abmmhasan
committed
41a03ba
docs(ai): document FastFlow Graphify OpenAI path
abmmhasan
committed
81b4044
docs(plan): finalize provider-aware Graphify flow
abmmhasan
committed
89fc7e4
refactor(ai): hardcode latest tags and common think switch
abmmhasan
committed
143b03e
refactor(ai): remove obsolete LLM image architecture selector
abmmhasan
committed
c69e509
refactor(ai): select Ollama tags without LDS_LLM_ARCH
abmmhasan
committed
a6e3089
feat(env): add dotenv key removal helper
abmmhasan
committed
24b4b77
fix(ai): unify thinking controls and harden local Graphify
abmmhasan
committed
06f9557
feat(ai): add provider-neutral thinking switch
abmmhasan
committed
ec1da99
refactor(ai): drop LDS_LLM_ARCH migration references
abmmhasan
committed
0e6773e
feat(ai): expose provider-neutral thinking setting to consumers
abmmhasan
committed
3e3f389
feat(ai): seed common thinking override
abmmhasan
committed
d6517b6
test(ai): lock common thinking and direct image tag contracts
abmmhasan
committed
3fe0d2b
docs(ai): document common thinking and direct image tags
abmmhasan
committed
80bd8ce
refactor(setup): remove obsolete LLM architecture env
abmmhasan
committed
19ad7cb
fix(graphify): preserve external target semantics
abmmhasan
committed
575c6f1
test(compose): align direct Ollama image tag contract
abmmhasan
committed
0f5df7e
test(images): align direct Ollama tag selection
abmmhasan
committed
98abe4c
docs(plan): remove obsolete LLM architecture selector
abmmhasan
committed
110a16d
test(env): remove obsolete LLM architecture contract
abmmhasan
committed
70f267d
fix(catalog): pair common thinking default with setup prompt
abmmhasan
committed
752f651
test(catalog): allow automatic AI thinking default
abmmhasan
committed
8e10863
fix(images): use direct LLM image tags
abmmhasan
committed
8a18bea
test(docs): assert provider-aware no-thinking Graphify contract
abmmhasan
committed
5f0ecb1
docs(ai): remove obsolete LLM tag variable wording
abmmhasan
committed
f100ef2
docs(ai): remove obsolete LLM tag variable reference
abmmhasan
committed
c71a210
feat(graphify): add suspect-response diagnostic proxy
abmmhasan
committed
997c474
fix(graphify): inspect extraction responses only
abmmhasan
committed
4f86d35
feat(graphify): capture suspect local LLM responses
abmmhasan
committed
310fa8f
test(graphify): cover suspect-response diagnostics
abmmhasan
committed
39db7d1
fix(graphify): bound non-text diagnostic responses
abmmhasan
committed
04b1214
fix(graphify): resolve diagnostic Python safely
abmmhasan
committed
efcdc88
docs(graphify): explain suspect-response diagnostics
abmmhasan
committed
33a3edb
test(docs): lock Graphify diagnostic guidance
abmmhasan
committed
9a83b56
fix(graphify): classify malformed graph JSON explicitly
abmmhasan
committed
d94d479
test(graphify): identify malformed graph JSON
abmmhasan
committed
82ba0ac
feat(graphify): add thinking A/B override
abmmhasan
committed
08ae9a3
test(graphify): cover thinking A/B override
abmmhasan
committed
ab58b6f
docs(graphify): document thinking A/B test
abmmhasan
committed
438e688
feat(graphify): recover malformed FastFlow graphs via tool calling
abmmhasan
committed
8545379
test(graphify): cover tool-call recovery
abmmhasan
committed
45a2e59
docs(graphify): explain structured FastFlow recovery
abmmhasan
committed
279c645
fix(graphify): constrain recovery tool to Graphify schema
abmmhasan
committed
3c76770
perf(graphify): make structured FastFlow extraction primary
abmmhasan
committed
7f4b460
fix(graphify): scope structured primary path to FastFlow
abmmhasan
committed
0a67590
feat(graphify): use native structured extraction for both providers
abmmhasan
committed
a748baf
fix(graphify): keep schema below future import
abmmhasan
committed
b7b146b
fix(graphify): make structured extraction provider-explicit
abmmhasan
committed
df0ad4c
fix(graphify): pass active provider to structured proxy
abmmhasan
committed
3f81e61
test(graphify): cover provider-native structured extraction
abmmhasan
committed
7ae08f9
refactor(graphify): decouple structured proxy from diagnostics
abmmhasan
committed
ecd0faa
refactor(graphify): keep structured proxy always enabled
abmmhasan
committed
ea8c1ce
test(graphify): lock always-on structured proxy contract
abmmhasan
committed
8ccbfb0
docs(graphify): document provider-native structured extraction
abmmhasan
committed
817c7e4
fix(graphify): align edge relations with Graphify schema
abmmhasan
committed
d399cfe
test(docs): align Graphify structured-output guidance
abmmhasan
committed
f010d1f
fix(graphify): bound structured FastFlow retries
abmmhasan
committed
e773a33
feat(graphify): expose bounded structured timeout
abmmhasan
committed
d84beec
test(graphify): cover bounded structured retry
abmmhasan
committed
7afb7ae
docs(graphify): document bounded structured extraction
abmmhasan
committed
5bc906b
test(graphify): lock structured timeout contract
abmmhasan
committed
dc57dbe
feat(graphify): surface incremental update mode
abmmhasan
committed
d0d21e9
fix(graphify): keep incremental mode detection minimal
abmmhasan
committed
1c47a60
test(graphify): lock incremental update mode detection
abmmhasan
committed
ac053a7
docs(graphify): clarify incremental existing-graph behavior
abmmhasan
committed
0226b15
test(docs): lock incremental Graphify guidance
abmmhasan
committed
23ff9e8
fix(graphify): fail fast and delegate chunk retries
abmmhasan
committed
058c638
fix(graphify): disable hidden SDK retry amplification
abmmhasan
committed
dacb72f
test(graphify): cover fail-fast structured retry delegation
abmmhasan
committed
db53a9e
docs(graphify): document fail-fast structured extraction
abmmhasan
committed
7ce29cf
fix(graphify): bound local adaptive retry depth
abmmhasan
committed
670851a
test(graphify): lock bounded adaptive retry depth
abmmhasan
committed
2e20973
docs(graphify): document bounded retry depth
abmmhasan
committed
bacee06
fix(graphify): preserve caller target for explicit providers
abmmhasan
committed
7c65690
fix(graphify): tolerate FastFlow tool-call argument shapes
abmmhasan
committed
622026e
test(graphify): cover FastFlow tool-call argument variants
abmmhasan
committed
e63924e
fix(graphify): preserve timeout split signal
abmmhasan
committed
01b1915
test(graphify): preserve timeout length signal
abmmhasan
committed
468dee7
fix(graphify): consume FastFlow tool calls from stream
abmmhasan
committed
4594fb5
test(graphify): cover streamed FastFlow tool completion
abmmhasan
committed
2c9eaa2
docs(graphify): document FastFlow streaming tool adapter
abmmhasan
committed
77a66eb