https://github.com/slartibardfast/yarn-agentic
· scanned 2026-06-16 00:15 UTC (2 months, 2 weeks ago)
55 raw signals (0 security + 55 graph)
Last scanned 2 months, 2 weeks ago · v1 · 54 actionable findings from 1 signal source. 1 repeated signal grouped for readability. Security checks, system graph analysis, and verified AI-agent feedback are merged into one review queue.
All 632 nodes from the latest scan, grouped by kind. Each node is a unit the engine identified (file, function, endpoint, table…). Most users won't need this view — it's primarily for debugging the engine's graph extraction or for AI agents that want to enumerate the project structure.
| Label | Layer | Status | Path |
|---|---|---|---|
README.md |
software | healthy | README.md |
COORD.md |
software | healthy | COORD.md |
CLAUDE.md |
software | healthy | CLAUDE.md |
book.toml |
software | healthy | book.toml |
benchmark.sh |
software | healthy | benchmark.sh |
STATUS.md |
software | healthy | STATUS.md |
PLAN.md |
software | healthy | PLAN.md |
MEMORY.md |
software | healthy | MEMORY.md |
compare_kv_quants.sh |
software | healthy | reference/ppl/compare_kv_quants.sh |
README.md |
software | healthy | reference/agner/README.md |
verify_projection.sh |
software | healthy | reference/agner/verify_projection.sh |
analyze.py |
software | healthy | reference/agner/analyze.py |
extract.py |
software | healthy | reference/agner/extract.py |
gate.yaml |
software | healthy | tests/mtp-fused/gate.yaml |
benchmark_async.sh |
software | healthy | benchmarks/benchmark_async.sh |
home.md |
software | healthy | docs/home.md |
SUMMARY.md |
software | healthy | docs/SUMMARY.md |
BENCHMARK_EXPECTATIONS.md |
software | healthy | docs/reference/BENCHMARK_EXPECTATIONS.md |
DISPATCH_COMPARISON.md |
software | healthy | docs/reference/DISPATCH_COMPARISON.md |
MULTI-GPU-PER-STEP-CHECKPOINT.md |
software | healthy | docs/reference/MULTI-GPU-PER-STEP-CHECKPOINT.md |
ENV_VARS.md |
software | healthy | docs/reference/ENV_VARS.md |
TU102_KERNEL_REVIEW_20260609.md |
software | healthy | docs/reference/TU102_KERNEL_REVIEW_20260609.md |
BENCHMARKS.md |
software | healthy | docs/reference/BENCHMARKS.md |
WSL2_MIRRORED_LOOPBACK.md |
software | healthy | docs/reference/WSL2_MIRRORED_LOOPBACK.md |
MTP_INSTRUCTIONS.md |
software | healthy | docs/reference/MTP_INSTRUCTIONS.md |
DATA.md |
software | healthy | docs/reference/DATA.md |
TRANSFER.md |
software | healthy | docs/reference/handoffs/TRANSFER.md |
HANDOFF.md |
software | healthy | docs/reference/handoffs/HANDOFF.md |
RESEARCH_2026-05-16.md |
software | healthy | docs/archive/refs-2026-05-16/RESEARCH_2026-05-16.md |
AUDIT_2026-05-16_CX_A_AFTERMATH.md |
software | healthy | docs/archive/refs-2026-05-16/AUDIT_2026-05-16_CX_A_AFTERMAT… |
PHASE36-MULTI-GPU-MTP-DRAFT-THROUGHPUT.md |
software | healthy | docs/archive/phases/multi-gpu/PHASE36-MULTI-GPU-MTP-DRAFT-T… |
PHASE41.md |
software | healthy | docs/archive/phases/multi-gpu/PHASE41.md |
PHASE19.md |
software | healthy | docs/archive/phases/multi-gpu/PHASE19.md |
PHASE38.md |
software | healthy | docs/archive/phases/multi-gpu/PHASE38.md |
PHASE40.md |
software | healthy | docs/archive/phases/multi-gpu/PHASE40.md |
PHASE20.md |
software | healthy | docs/archive/phases/multi-gpu/PHASE20.md |
PHASE22.md |
software | healthy | docs/archive/phases/multi-gpu/PHASE22.md |
PHASE44_SYNTHESIS.md |
software | healthy | docs/archive/phases/multi-gpu/PHASE44_SYNTHESIS.md |
PHASE7.md |
software | healthy | docs/archive/phases/multi-gpu/PHASE7.md |
PHASE9.md |
software | healthy | docs/archive/phases/multi-gpu/PHASE9.md |
PHASE39-TREE-DRAFTING.md |
software | healthy | docs/archive/phases/multi-gpu/PHASE39-TREE-DRAFTING.md |
PHASE17.md |
software | healthy | docs/archive/phases/multi-gpu/PHASE17.md |
PHASE2.md |
software | healthy | docs/archive/phases/multi-gpu/PHASE2.md |
PHASE36-CLOSURE.md |
software | healthy | docs/archive/phases/multi-gpu/PHASE36-CLOSURE.md |
PHASE10.md |
software | healthy | docs/archive/phases/multi-gpu/PHASE10.md |
PHASE46-MULTIGPU-CLIP-TENSOR-SPLIT.md |
software | healthy | docs/archive/phases/multi-gpu/PHASE46-MULTIGPU-CLIP-TENSOR-… |
PHASE36-FUSED-DESIGN-TRACE.md |
software | healthy | docs/archive/phases/multi-gpu/PHASE36-FUSED-DESIGN-TRACE.md |
PHASE21.md |
software | healthy | docs/archive/phases/multi-gpu/PHASE21.md |
PHASE44.md |
software | healthy | docs/archive/phases/multi-gpu/PHASE44.md |
PHASE36-PLAN.md |
software | healthy | docs/archive/phases/multi-gpu/PHASE36-PLAN.md |
Showing first 50 of this kind. Full payload available via the JSON button at the top of the page.
| Label | Layer | Status | Path |
|---|---|---|---|
load_recip |
software | healthy | reference/agner/analyze.py:35 |
main |
software | healthy | reference/agner/analyze.py:58 |
cell_text |
software | healthy | reference/agner/extract.py:125 |
iter_sheet_rows |
software | healthy | reference/agner/extract.py:131 |
detect_columns |
software | healthy | reference/agner/extract.py:146 |
norm_operand |
software | healthy | reference/agner/extract.py:176 |
find_row |
software | healthy | reference/agner/extract.py:179 |
main |
software | healthy | reference/agner/extract.py:196 |
get |
software | healthy | reference/agner/extract.py:221 |
build_prompt |
software | healthy | docs/archive/phases/qwen35-tool-calling/profile/mtp_ctx_swe… |
run_one |
software | healthy | docs/archive/phases/qwen35-tool-calling/profile/mtp_ctx_swe… |
main |
software | healthy | docs/archive/phases/qwen35-tool-calling/profile/mtp_ctx_swe… |
parse |
software | healthy | docs/archive/phases/qwen35-tool-calling/profile/parse_vk_pe… |
main |
software | healthy | docs/archive/phases/qwen35-tool-calling/profile/parse_vk_pe… |
run_one |
software | healthy | docs/archive/phases/qwen35-tool-calling/profile/drive.py:38 |
main |
software | healthy | docs/archive/phases/qwen35-tool-calling/profile/drive.py:59 |
load_drive |
software | healthy | docs/archive/phases/qwen35-tool-calling/profile/vega-vs-nav… |
main |
software | healthy | docs/archive/phases/qwen35-tool-calling/profile/vega-vs-nav… |
load_perf |
software | healthy | docs/archive/phases/qwen35-tool-calling/profile/fusion-ab/b… |
load_drive |
software | healthy | docs/archive/phases/qwen35-tool-calling/profile/fusion-ab/b… |
main |
software | healthy | docs/archive/phases/qwen35-tool-calling/profile/fusion-ab/b… |
parse |
software | healthy | docs/archive/phases/qwen35-tool-calling/profile/op-coverage… |
main |
software | healthy | docs/archive/phases/qwen35-tool-calling/profile/op-coverage… |
parse_stderr |
software | healthy | docs/archive/phases/qwen35-tool-calling/profile/vram-budget… |
main |
software | healthy | docs/archive/phases/qwen35-tool-calling/profile/vram-budget… |
parse_startup |
software | healthy | docs/archive/phases/qwen35-tool-calling/profile/kv-matrix/b… |
parse_drive |
software | healthy | docs/archive/phases/qwen35-tool-calling/profile/kv-matrix/b… |
parse_perf |
software | healthy | docs/archive/phases/qwen35-tool-calling/profile/kv-matrix/b… |
main |
software | healthy | docs/archive/phases/qwen35-tool-calling/profile/kv-matrix/b… |
load_schemas |
software | healthy | docs/archive/phases/qwen35-tool-calling/harness/run_eval.py… |
load_cases |
software | healthy | docs/archive/phases/qwen35-tool-calling/harness/run_eval.py… |
build_tool |
software | healthy | docs/archive/phases/qwen35-tool-calling/harness/run_eval.py… |
_http_post |
software | healthy | docs/archive/phases/qwen35-tool-calling/harness/run_eval.py… |
build_request_openai |
software | dead | docs/archive/phases/qwen35-tool-calling/harness/run_eval.py… |
parse_response_openai |
software | dead | docs/archive/phases/qwen35-tool-calling/harness/run_eval.py… |
build_request_hermes |
software | dead | docs/archive/phases/qwen35-tool-calling/harness/run_eval.py… |
parse_response_hermes |
software | dead | docs/archive/phases/qwen35-tool-calling/harness/run_eval.py… |
_matches_json_type |
software | healthy | docs/archive/phases/qwen35-tool-calling/harness/run_eval.py… |
score_case |
software | healthy | docs/archive/phases/qwen35-tool-calling/harness/run_eval.py… |
aggregate |
software | healthy | docs/archive/phases/qwen35-tool-calling/harness/run_eval.py… |
rate |
software | healthy | docs/archive/phases/qwen35-tool-calling/harness/run_eval.py… |
run |
software | healthy | docs/archive/phases/qwen35-tool-calling/harness/run_eval.py… |
main |
software | healthy | docs/archive/phases/qwen35-tool-calling/harness/run_eval.py… |
is_mtp_core_file |
software | healthy | mtp-extract/classify.py:21 |
is_mtp_test_file |
software | healthy | mtp-extract/classify.py:52 |
is_mtp_doc_signal |
software | healthy | mtp-extract/classify.py:55 |
subject_is_out_scope |
software | healthy | mtp-extract/classify.py:71 |
file_is_out_of_scope |
software | healthy | mtp-extract/classify.py:75 |
file_is_mtp_signal |
software | healthy | mtp-extract/classify.py:89 |
main |
software | healthy | mtp-extract/classify.py:94 |
Showing first 50 of this kind. Full payload available via the JSON button at the top of the page.
| Label | Layer | Status | Path |
|---|---|---|---|
reference |
software | healthy | reference |
ppl |
software | healthy | reference/ppl |
agner |
software | healthy | reference/agner |
tests |
software | healthy | tests |
mtp-fused |
software | healthy | tests/mtp-fused |
benchmarks |
software | healthy | benchmarks |
docs |
software | healthy | docs |
reference |
software | healthy | docs/reference |
handoffs |
software | healthy | docs/reference/handoffs |
archive |
software | healthy | docs/archive |
refs-2026-05-16 |
software | healthy | docs/archive/refs-2026-05-16 |
phases |
software | healthy | docs/archive/phases |
multi-gpu |
software | healthy | docs/archive/phases/multi-gpu |
quantization |
software | healthy | docs/archive/phases/quantization |
kernels-and-scheduling |
software | healthy | docs/archive/phases/kernels-and-scheduling |
qwen35-tool-calling |
software | healthy | docs/archive/phases/qwen35-tool-calling |
profile |
software | healthy | docs/archive/phases/qwen35-tool-calling/profile |
vega-vs-navi |
software | healthy | docs/archive/phases/qwen35-tool-calling/profile/vega-vs-navi |
fusion-ab |
software | healthy | docs/archive/phases/qwen35-tool-calling/profile/fusion-ab |
op-coverage |
software | healthy | docs/archive/phases/qwen35-tool-calling/profile/op-coverage |
mtp-sweep |
software | healthy | docs/archive/phases/qwen35-tool-calling/profile/mtp-sweep |
per-op |
software | healthy | docs/archive/phases/qwen35-tool-calling/profile/per-op |
vram-budget |
software | healthy | docs/archive/phases/qwen35-tool-calling/profile/vram-budget |
pipeline-stats |
software | healthy | docs/archive/phases/qwen35-tool-calling/profile/pipeline-st… |
kv-matrix |
software | healthy | docs/archive/phases/qwen35-tool-calling/profile/kv-matrix |
harness |
software | healthy | docs/archive/phases/qwen35-tool-calling/harness |
results |
software | healthy | docs/archive/phases/qwen35-tool-calling/harness/results |
schemas |
software | healthy | docs/archive/phases/qwen35-tool-calling/harness/schemas |
determinism |
software | healthy | docs/archive/phases/determinism |
infrastructure |
software | healthy | docs/archive/phases/infrastructure |
speculative-decoding |
software | healthy | docs/archive/phases/speculative-decoding |
theme |
software | healthy | docs/theme |
active |
software | healthy | docs/active |
mtp-extract |
software | healthy | mtp-extract |
scripts |
software | healthy | scripts |
systemd |
software | healthy | scripts/systemd |
gpu-clocks |
software | healthy | scripts/systemd/gpu-clocks |
llm-rt-tuning |
software | healthy | scripts/systemd/llm-rt-tuning |
policy |
software | healthy | scripts/policy |
overnight-soak |
software | healthy | scripts/overnight-soak |
kdump |
software | healthy | scripts/kdump |
cuda-graph-probe |
software | healthy | scripts/cuda-graph-probe |
services |
software | healthy | services |
anthropic-proxy |
software | healthy | services/anthropic-proxy |
test |
software | healthy | services/anthropic-proxy/test |
.github |
software | healthy | .github |
workflows |
software | healthy | .github/workflows |
specs |
software | healthy | specs |
multislot |
software | healthy | specs/multislot |
dispatch |
software | healthy | specs/dispatch |
Showing first 50 of this kind. Full payload available via the JSON button at the top of the page.
| Label | Layer | Status | Path |
|---|---|---|---|
port:18181 |
network | healthy | scripts/_check-output-parity.sh |
port:18292 |
network | healthy | scripts/test-production-np-determinism.sh |
port:8080 |
network | healthy | scripts/repro-state-restore-cascade.sh |
port:18080 |
network | healthy | scripts/test-pp-serialization.sh |
port:07 |
network | healthy | scripts/deploy-llama-server.sh |
port:52 |
network | healthy | scripts/deploy-llama-server.sh |
port:10 |
network | healthy | scripts/deploy-llama-server.sh |
port:27 |
network | healthy | scripts/deploy-llama-server.sh |
port:37 |
network | healthy | scripts/deploy-llama-server.sh |
port:18291 |
network | healthy | scripts/test-fattn-per-slot-kv-np-determinism.sh |
port:18290 |
network | healthy | scripts/test-mtp-multislot-determinism.sh |
port:18293 |
network | healthy | scripts/verify-multigpu-clip.sh |
| Label | Layer | Status | Path |
|---|---|---|---|
Policy |
software | healthy | scripts/recast_bf16_to_fp16.py:47 |
TensorAction |
software | healthy | scripts/recast_bf16_to_fp16.py:155 |
Violation |
software | healthy | scripts/validate-batch-composition-trace.py:106 |
ValidationError |
software | healthy | scripts/validate-paged-allocator-trace.py:52 |
Handler |
software | healthy | services/anthropic-proxy/test/mock_openai.py:20 |
| Label | Layer | Status | Path |
|---|---|---|---|
127.0.0.1 |
network | healthy | docs/archive/phases/qwen35-tool-calling/profile/mtp_ctx_swe… |
127.0.0.2 |
network | healthy | scripts/deploy-llama-server.sh |
| Label | Layer | Status | Path |
|---|---|---|---|
gha::pages |
cicd | healthy | .github/workflows/pages.yml |
gha::spec-tla-gate |
cicd | healthy | .github/workflows/spec-tla-gate.yml |
| Label | Layer | Status | Path |
|---|---|---|---|
deploy |
cicd | healthy | .github/workflows/pages.yml |
spec-tla |
cicd | healthy | .github/workflows/spec-tla-gate.yml |
| Label | Layer | Status | Path |
|---|---|---|---|
repobility-clone-ge5d0s2p |
software | healthy | /tmp/repobility-clone-ge5d0s2p |
| Label | Layer | Status | Path |
|---|---|---|---|
nginx |
network | healthy | scripts/snoop-summarise.py |
| Label | Layer | Status | Path |
|---|---|---|---|
gpu (detected) |
hardware | healthy | README.md |
| Label | Layer | Status | Path |
|---|---|---|---|
vps::ec2 |
hardware | healthy | scripts/test-pp-serialization.sh |
This page is publicly accessible at:
https://repobility.com/scan/9e5912e0-46bc-4281-a283-9f77bfa71fc6/
To check status programmatically (no auth required):
curl -s https://repobility.com/api/v1/public/scan/9e5912e0-46bc-4281-a283-9f77bfa71fc6/
Important — please don't re-submit the same URL repeatedly. The submission endpoint is idempotent: re-submitting the same git URL returns this same scan_token, not a new one. To re-scan this repo, sign up free and use the dashboard.