[codex] add CERC plan validation#3
Draft
milkoor wants to merge 9 commits into
Draft
Conversation
Add read-only corpus phase4-status subcommand that reports all 8 Phase 4-3 trigger conditions with current values, thresholds, met/not-met status, affected theory candidates, and next actions. All metrics scoped correctly: Trigger 7 (runtime balance) counts only native strict sessions. Triggers 3/4 count tagged intervention lane sessions. Output format: table with current value, threshold, and status, followed by summary and per-trigger action guidance. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Conservative classification using only explicit metadata fields (data_origin, task_source, intervention_lane, causetrace_tags). High confidence: data_origin=native + task_source=real_work. Medium confidence: data_origin=native with no task_source. Dry-run only — no metadata writes. Tracks skip reasons for all unmatched sessions. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Writes intervention_lane=direct_prompt_native for high-confidence proposals only (data_origin=native + task_source=real_work + no intervention markers). Medium-confidence proposals are never applied. Adds provenance: classified_from_explicit_metadata. Fixes native_strict counting in phase4-status to treat direct_prompt_native as the native baseline lane. Before: 12/992 classified (1.2%) After: 113/992 classified (11.4%) 879 unknown sessions remain unmodified. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Documents current state after A2 classification pass: - 108/992 sessions with explicit intervention_lane (10.9%) - 101 direct_prompt_native via classified_from_explicit_metadata - 879 data_origin=unknown sessions cannot be auto-classified Defines upstream capture requirements (P0-P5), source-specific defaults, capture points, and explicit non-goals. Trigger 8 should improve through future capture quality, not aggressive retroactive inference. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Operational checklists for each capture type: - direct native real work - routed prompt task - superpowers workflow - controlled prompt morphology pilot - external trajectory - demo/test run Includes post-capture verification commands and explicit non-rules (runtime-only, Skill-only, prompt-style inference are not valid). Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
These were committed before the ignore rules were added. Now .gitignore will prevent them from being re-added. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What changed
causetrace corpus validate-planto check CERC experiment plans for duplicate signatures, queue constraint validity, and sampling necessity.causetrace.crdd.plan_validationplus a read-only research note for the new validation layer.Why
Impact
~/.causetrace/plan_validation.Validation
python3 -m pytest tests/test_metadata_corpus_report.py -v -k "plan_validation or cerc_feedback"python3 -m pytest tests/ -v_handle_corpusbefore the CLI edit.