Fetching metadata from the HF Docker repository...
Sep 28 10:25 Structured arguments items 1-2: thesis rests_on + two-pass gate with report context, one bounded repair turn, typed trigger labels, evidence printed in full once, identifier and authored-once audits, relaxed evidence limits, past returns withheld Sep 23 19:41 ML models: walk-forward-judged panel rank, plain-language ML disclosure in report and progress 19:01 Storytellers run in parallel; 'under the hood' progress narration of Damodaran's framework inputs 18:40 progress: story-start marker, updated timings; summary limited to two figures 18:32 Summary-first report with expandable sections, mobile UI, live progress (storytellers, debate seats), low-risk stable-rate override 09:22 Failed runs mirror their log tail to the private QA dataset 09:13 Engine log tail in the private QA record; Damodaran replay tool; stable-growth sensitivity hook Sep 22 20:24 AnalysisState pipeline: reconciliation, staged validation, frozen state, claim-token prose, audited renderer; public/QA split 19:43 Privacy: requester IP, user-agent and referer move to a private server log; reports scrubbed on save, load and serve; committed FRED key removed 10:38 P0 hardening: one percentage type, exhaustive rendered-claim audit, engine-owned terminology, reliability as evidence permissions, typed metric categories, FPI leadership, 10b5-1 semantics, deterministic ranges 07:48 Semantic types: normalized metrics (currency/scale/ADS basis) with gated derivations; engine-owned labels and methodology; one ScenarioSet and immutable object ids; reliability controls weight, language and persona visibility; structured thesis triggers with causal polarity; separate intrinsic/pricing/momentum signals; claim-registry render audit; 8 publication gates; 95 tests 06:45 Module-level ERP constant; story gate reads the policy; non-FCFF engines set the intrinsic value and label; pass segment/bank/ADS facts through 05:21 Architecture: typed concepts + immutable AnalysisContext with rejected-field quarantine and lineage; applicability policy; valuation engines with label invariants; quant reliability states; semantic comparisons; engine-side synthesis/scenario EV; five-state checks; six publication gates incl. final render audit; property/golden/corruption tests Sep 21 12:57 Pipeline remediation P0: canonical layer (issuer typing, framework selection, metric lineage, SEC XBRL cross-checks, Stage A validation), consensus-availability handling, story bridge/dispersion, corporate-action state machine, quote/return-kind rules, two-stage publication gate, guardrail tests 11:53 Report-level validation gate: canonical rating, reproducible blend/composite audits, canonical PEG, Form 4 insider context, SEC-first leadership, mechanical momentum, REIT/services split, markdown export 10:40 Upload analyst.py with huggingface_hub 10:24 Upload analyst.py with huggingface_hub 10:18 Upload analyst.py with huggingface_hub 09:01 Upload analyst.py with huggingface_hub 05:20 Upload analyst.py with huggingface_hub Sep 7 05:17 Upload analyst.py with huggingface_hub Sep 3 13:31 Upload analyst.py with huggingface_hub 13:02 Upload analyst.py with huggingface_hub 10:34 Upload analyst.py with huggingface_hub 09:13 Upload analyst.py with huggingface_hub 09:03 Upload analyst.py with huggingface_hub Aug 31 08:44 Upload analyst.py with huggingface_hub Aug 27 05:26 Upload analyst.py with huggingface_hub 05:20 Upload analyst.py with huggingface_hub Aug 25 10:36 Upload analyst.py with huggingface_hub Aug 23 13:48 Upload analyst.py with huggingface_hub 13:41 Upload analyst.py with huggingface_hub Aug 17 11:40 Upload analyst.py with huggingface_hub Aug 13 12:35 Upload analyst.py with huggingface_hub 06:18 Upload analyst.py with huggingface_hub Aug 9 08:54 Upload analyst.py with huggingface_hub Aug 6 13:38 Upload analyst.py with huggingface_hub 12:30 Crucible — AI equity analyst (containerized engine), secrets-only 
