Review copper heap and Coscinodiscus causal graphs (batch 25)

history/records/Copper_Biomining_Heap_Leach/2026-10-06T025228Z-codex-9c8b7f.yaml

@@ -0,0 +1,30 @@
+history_version: 1
+target:
+  kind: record
+  path: kb/communities/Copper_Biomining_Heap_Leach.yaml
+  slug: Copper_Biomining_Heap_Leach
+session:
+  id: 2026-10-06T025228Z-codex-9c8b7f
+  timestamp: '2026-10-06T02:52:28Z'
+  actors:
+  - type: ai_agent
+    name: codex
+    model: unspecified
+    agent_tool: codex
+links:
+  issues:
+  - https://github.com/CultureBotAI/CommunityMech/issues/1356
+events:
+- type: EDIT
+  outcome: changed
+  sections:
+  - interactions
+  - discussions
+  summary: Separate measured intrinsic iron oxidation from early-heap association and failed
+    recovery under SX reagent.; Retain observed succession with correct abundance scale and
+    species boundaries, without competition or resolved field flux.; Distinguish sulfur-oxidation
+    capacity and RNA activity from stable DNA counts, constant acid flux and selective passivation
+    removal.; Retain conditional mineral/redox feedback and direct laboratory outcomes without
+    a closed five-member consortium claim.; Preserve laboratory Sulfobacillus oxidation/leaching
+    and low-ORP advantage; bound field genus-to-species and thermal inference.
+  details: Source-based self-adversarial review; no new nodes/arrows. reports/causal_graph_review/decisions/20261005-two-records-batch25.yaml

history/records/Coscinodiscus_Synthetic_Community/2026-10-06T025228Z-codex-c5e0cf.yaml

@@ -0,0 +1,31 @@
+history_version: 1
+target:
+  kind: record
+  path: kb/communities/Coscinodiscus_Synthetic_Community.yaml
+  slug: Coscinodiscus_Synthetic_Community
+session:
+  id: 2026-10-06T025228Z-codex-c5e0cf
+  timestamp: '2026-10-06T02:52:28Z'
+  actors:
+  - type: ai_agent
+    name: codex
+    model: unspecified
+    agent_tool: codex
+links:
+  issues:
+  - https://github.com/CultureBotAI/CommunityMech/issues/1357
+events:
+- type: EDIT
+  outcome: changed
+  sections:
+  - interactions
+  - discussions
+  summary: Retain candidate diatom resource provision and actual Crocei1 promotion without
+    universal dependence or measured metabolite donation.; Keep DMSP homology potential computational,
+    distinct from consumption, product flux and a negative carbon-only growth assay.; Preserve
+    distinct reciprocal growth signs and control-relative algal recovery rather than global
+    mutualism.; Preserve real pooled nutrient responses and the CS4 negative control without
+    individual-vitamin rescue or exact B12-form claims.; Retain empirical community inhibition
+    and phase-specific model success without reciprocal harm or universal exclusion of higher-order
+    interactions.
+  details: Source-based self-adversarial review; no new nodes/arrows. reports/causal_graph_review/decisions/20261005-two-records-batch25.yaml

reports/causal_graph_review/adversarial-review-20261005-batch25.yaml

@@ -0,0 +1,130 @@
+reviewer: codex
+review_mode: source_based_self_adversarial_review
+independent_approval: false
+base_commit: 2eb1818f3d94f9420099a991136254501b543891
+decision_ledger: reports/causal_graph_review/decisions/20261005-two-records-batch25.yaml
+validation_report: reports/causal_graph_review/validation-20261005-batch25.json
+scope:
+- kb/communities/Copper_Biomining_Heap_Leach.yaml
+- kb/communities/Coscinodiscus_Synthetic_Community.yaml
+findings:
+- issue: 1356
+  severity: P2
+  finding: Field DNA/RNA succession promoted into resolved flux, competition and a closed syntrophic cycle; laboratory
+    conditions and species conflated.
+  disposition: addressed_and_locally_verified
+- issue: 1357
+  severity: P2
+  finding: DMSP homology and pooled nutrient rescue promoted into specific vitamin donation, chemical flux and global
+    ecological fitness signs.
+  disposition: addressed_with_documented_reference_cache_limit
+- issue: 1360
+  severity: P2
+  finding: Legitimate intrinsic-source corrections exceed the previous mixed-scope census ceiling of 122.
+  disposition: Exact census verified; only ceiling/comment changed to 124; full regression rerun passes and initial
+    failure remains recorded.
+open_nongraph_followups:
+- issue: 1358
+  disposition: not_closed_by_graph_only_patch
+- issue: 1359
+  disposition: not_closed_by_graph_only_patch
+adversarial_checks:
+- Every original node and arrow reviewed against whole records and cited primary sources. Retain ten nodes/ten arrows
+  with seven exact renames; no new structure or Edison job.
+- Copper field DNA/rRNA supports abundance/activity, not species-resolved oxidation flux or selective competition.
+  Ferroplasma approximately 10^3 copies/mL differs from late L. ferriphilum approximately 10^5.
+- The indigenous 2009 heap is an open community with a 400-day cycle and mixed copper sulfides, not a closed five-member
+  culture. Preserve conditional substrate/regeneration feedback, not unsupported syntrophy or niche partitioning.
+- Separate 2018 A. caldus/L. ferriphilum/S. thermosulfidooxidans cultures at 38 +/- 2 C from 2022 monocultures including
+  L. ferrooxidans, and from field genus-level Sulfobacillus RNA. Remove iron-ion-transport overannotation.
+- 'Preserve measured lower-redox leaching advantage and nonmonotonic response: A. ferrooxidans retains iron oxidation
+  under SX reagent despite copper recovery falling to abiotic level; Sulfobacillus low-dose recovery differs from
+  high-dose impairment.'
+- Stable A. thiooxidans DNA counts do not measure constant acid flux; RNA supports activity. General chemistry quoted
+  from the 2022 Introduction is OTHER/PARTIAL, not an in vivo five-member assay.
+- Coscinodiscus has real pooled amino-acid/vitamin supplementation assays. Preserve the three responders and CS4
+  negative control. Do not substitute a genome-only account, individual B12/B3 rescue or demonstrated diatom donation.
+- DMSP pathway BLASTP is computational homology with provenance, not measured consumption or DMS release. Carbon-only
+  supplementation failure does not prove no catabolism.
+- Remove unsupported cob(I)alamin exact-form assignment from the pooled supplement node while retaining specific
+  genome predictions in bounded prose. Use existing amino-acid and vitamin-role annotations for the tested mixture.
+- Preserve Crocei1 promotion, CS1 suppression, CS4/Rose1 neutral bacterial effects, and control-relative algal recovery.
+  Do not assert one global mutualism/commensalism or reciprocal competition sign.
+- Retain empirical all-member inhibition with late-survival nuance and generalized Lotka-Volterra model success
+  using separate days 1-17 and 17-41 parameter sets. No bacterial-bacterial terms needed for this fit does not prove
+  universal absence of higher-order interactions.
+- Three brief Results quotes total 15 words. They were verified in the ASM-provided primary preview and remain absent
+  from the abstract-only cache. Reference exit 1 and three MISMATCH diagnostics are preserved, not mislabeled passed
+  or repaired by circular cache text.
+- Four complete primary main bodies and three contextual abstracts reviewed. No figure images or supplementary files
+  inspected, no visually validated OCR equations, no new full copyrighted body or PDF in git.
+- 'Eight tracked cache artifacts, canonical taxonomy and unrelated raw YAML blocks preserved. Metadata followups
+  #1358/#1359 and broader cache/link issues remain open.'
+- 'Four intrinsic source-only scopes move four taxa from solely community credit while total canonical/community
+  coverage remains 1499. Both records become mixed-scope. Initial regressions: 292 pass and one stale ceiling failure.
+  #1360 changes only the ceiling 122 -> 124 and its explanatory comment; all other bounds, participant checks and
+  connectivity behavior remain unchanged. The complete suite is rerun successfully.'
+- Coverage is 123/460, not completion. This is source-based self-adversarial review, not independent approval or
+  protected CI success.
+verification:
+  status: complete_with_documented_reference_retrieval_gaps
+  checks: 27
+  passing_command_exits: 25
+  failed_reference_commands: 1
+  records: 460
+  nodes: 1474
+  edges: 710
+  review_status_counts:
+    needs_research: 3
+    pending: 334
+    reviewed: 123
+  structural_defects: 0
+  network_errors: 0
+  network_warnings: 25
+  new_network_warnings: 0
+  unchanged_raw_blocks: 27
+  snippet_counts:
+    MATCH: 46
+    RENDERING: 0
+    ASSEMBLED: 0
+    WEAK: 0
+    MISMATCH: 3
+    NOCONTENT: 0
+  published_description_coverage:
+    pages: 456
+    node_descriptions: 1460
+    edge_descriptions: 702
+  render_determinism: passed
+  regressions: 293 passed in 187.32s (0:03:07)
+  rights: 3 passed in 1.93s
+  rights_import_root_verified: true
+  caches_unchanged: true
+  primary_artifact_hashes_rechecked: true
+  corpus_sentinels_unchanged: false
+  connectivity_survey_before:
+    records: 460
+    with_community_level: 389
+    mixed: 122
+    community_level_only: 267
+    taxa: 1499
+    credited_solely_by_the_rule: 1221
+  connectivity_survey:
+    records: 460
+    with_community_level: 389
+    mixed: 124
+    community_level_only: 265
+    taxa: 1499
+    credited_solely_by_the_rule: 1217
+  connectivity_delta:
+    records: 0
+    with_community_level: 0
+    mixed: 2
+    community_level_only: -2
+    taxa: 0
+    credited_solely_by_the_rule: -4
+  main_commands: 23
+  remediation_commands: 4
+  initial_regression_failure_fixed: true
+  unresolved_failed_commands: 1
+  other_corpus_sentinels_unchanged: true
+  census_bound_issue: 1360