fix: Add UTF-8 encoding for markdown file operations by datawranglerai · Pull Request #1 · PlanExeOrg/PlanExe

datawranglerai · 2025-02-10T15:43:12Z

Fixed UnicodeEncodeError in run_plan_pipeline.py by:

Using UTF-8 encoding when writing SWOT analysis markdown file
Using UTF-8 encoding when reading SWOT markdown in ExpertReviewTask
Switched from luigi.LocalTarget.open() to built-in open() for encoding support

This resolves character encoding issues when handling non-ASCII text in markdown files.

Fixed UnicodeEncodeError in run_plan_pipeline.py by: - Using UTF-8 encoding when writing SWOT analysis markdown file - Using UTF-8 encoding when reading SWOT markdown in ExpertReviewTask - Switched from luigi.LocalTarget.open() to built-in open() for encoding support This resolves character encoding issues when handling non-ASCII text in markdown files.

neoneye · 2025-02-10T15:45:22Z

Thank you @datawranglerai

Phase 1 (Critical Security): - Fix SECRET_KEY validation to detect both 'your-secret-key' AND 'dev-secret-key' defaults - Fail hard in production (when FLASK_ENV=production or PLANEXE_PUBLIC_BASE_URL set) - Add session cookie security flags (SECURE, HTTPONLY, SAMESITE=Lax) - Update .env examples with SECRET_KEY generation command Phase 2 (Error Handling & UX): - Wrap OAuth callback in try/except for better error handling - Add profile field validation with clear error messages - Log warning when OAuth profile missing email - Update login.html to display error messages Addresses Issues #1, #3, #5, #6, #7 from OAUTH_ANALYSIS.md

… progress Unlike plan_retry which clears all artifacts and restarts from scratch, plan_resume preserves completed pipeline outputs and restores the run directory from the stored zip snapshot. Luigi then skips completed tasks and picks up where the pipeline left off. This is the #1 blocker for autonomous agent workflows (Proposal 87) and a prerequisite for the autonomous prompt optimization loop (Proposals 94/59), which needs fast re-runs that skip completed tasks. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

The enrich step asks the LLM to generate description and conflict_text for each lever, but the batch prompt only provided lever_id, name, and options. The consequences (what happens when the lever is pulled) and review (the documented primary trade-off) fields were loaded but never forwarded to the prompt, forcing the model to infer them from the name alone. This adds both fields to the per-lever details in the batch prompt so the LLM can produce grounded descriptions and conflict analysis. Identified by analysis/53_enrich_potential_levers synthesis (direction #1). Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

… tighter targets Bundles all proven improvements from analyses 69-73 into one PR: 1. Verbatim-numbers constraint on BOTH consequences AND options: "Use numbers only when the project context provides them directly — do not calculate, derive, or estimate figures." Closes the arithmetic-derivation loophole (HK$470M→HK$141M). 2. Fix review_lever template lock (#1 issue from analysis 40): Replace "state the specific gap the three options leave unaddressed" with "identify the key risk or constraint that the proposed options collectively do not resolve." Removes "the three options" as grammatical anchor that caused 70-94% haiku template lock. 3. Positive framing: replace "Do NOT include 'Controls ... vs.'" with "Save critical assessments for the review_lever field." 4. Consistent targets everywhere: consequences 2-3 sentences, options one sentence, review_lever one sentence (20-40 words). All field descriptions and system prompt sections aligned. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

…n-physics plans The LLM instruction for checklist item #1 confused legal/treaty "laws" (constitutional law, EU treaty law) with "laws of physics", causing a Denmark euro adoption plan to be rated HIGH with mentions of FTL propulsion and perpetual motion. The instruction now explicitly distinguishes legal/regulatory/policy "laws" from physics laws and requires naming a specific physical law that is violated before rating above LOW. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

…+ MARGINAL bucketing ChatGPT v44 review: two wording fixes. Suggested next actions item #1 previously said 'N gate(s) currently fail at the 50% pass-rate bar' regardless of whether the worst pass rate was 0% or 49%. That phrasing understated DOOM failures: '1 gate fails at the 50% bar' reads identically whether the gate is FRAGILE-48% or DOOM-0%. Now distinguishes DOOM vs FRAGILE counts and names the worst gate by id + pass rate: '1 declared gate in the DOOM band. Worst: sponsor_profitability_window_margin_days at 0.0% pass rate under current bounds.' (or '2 in the DOOM band; 3 in the FRAGILE band. Worst: ... at X.X% pass rate.' for mixed cases). Decision implications MARGINAL wording was 'close enough to coin-flip' across the full 50-80% band. At 79.8% that reads as a misdiagnosis — the gate is one slip from ROBUST, not coin-flip. Bucketed at 70%: at-or-above 70% uses 'just below the ROBUST band. The gate passes in most runs, but downstream commitments should not treat it as secure.'; below 70% keeps the 'close to coin-flip' framing. No schema bump (manifest unchanged; pure rendering). Smoke 9/9, unit 50/50.

…hip-set Updates two docs to reflect the post-PlanExeOrg#753 state of the napkin-math pipeline. methology.md: describe the current pipeline behaviour — two-batch compress with paraphrase-tolerant quote match and cross-bucket promoter; extract's source-arithmetic preservation, threshold-pairing, and dropped_signals field; 19-check validator (added aggregate_not_bounded, requirement_has_margin, dropped_signals_schema); bounds' asymmetric source label on commitment defaults, calculation-output strip, reserved correlations block, reserved lognormal/pert disciplines with loud NotImplementedError; advisory audit_source_preservation.py step. 20260520_plan.md → 20260522_plan.md: bump status date; mark PR PlanExeOrg#750 merged; add PR PlanExeOrg#751/PlanExeOrg#752/PlanExeOrg#753 entries (proposal 141 implementation); update Phase status table (added 4.5 audit row, reclassified Phase 8 as partially done, Phase 10 marked done for current ship-set); add v58 14-plan empirical snapshot (1 viable / 5 fragile / 8 doom); reorder Next likely move now that proposal 141 has shipped — Phase 5 citation verifier promoted to PlanExeOrg#1, Phase 8 samplers added as PlanExeOrg#2 with v58 cases that bite now, Phase 9 composite-band cap as PlanExeOrg#3. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

neoneye merged commit a878286 into PlanExeOrg:mainFeb 10, 2025

82deutschmark mentioned this pull request Feb 8, 2026
OAuth Security Fixes: SECRET_KEY validation, session cookies, error handling #16
Merged

This was referenced Mar 9, 2026
Add plan_resume MCP tool for mid-pipeline recovery #217
Merged
Add validation error feedback for LLM retry attempts (#102) #221
Merged

neoneye mentioned this pull request Mar 29, 2026
Include consequences and review in enrich batch prompt #451
Merged
3 tasks

neoneye mentioned this pull request Apr 1, 2026
PR #479 + minimal review_lever to avoid template lock #482
Closed
5 tasks

neoneye mentioned this pull request Apr 16, 2026
fix: self-audit physics false positive on non-physics plans #581
Merged
2 tasks

neoneye mentioned this pull request May 17, 2026
feat(napkin-math): sharper severity wording in Suggested next actions + MARGINAL bucketing #722
Closed
2 tasks

This was referenced May 17, 2026
fix(napkin-math): saturated-gate note + DOOM severity wording + MARGINAL band bucketing #723
Merged
docs(napkin-math): refresh methodology + plan status for 2026-05-22 ship-set #754
Merged
napkin-math: wire prior-baseline ledger into Stage 0 reruns #756
Open

Navigation Menu

Uh oh!

fix: Add UTF-8 encoding for markdown file operations - #1

fix: Add UTF-8 encoding for markdown file operations#1
neoneye merged 1 commit into
PlanExeOrg:mainfrom
datawranglerai:fix/utf8-encoding-markdown

datawranglerai commented Feb 10, 2025

Uh oh!

neoneye commented Feb 10, 2025

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Uh oh!

Conversation

datawranglerai commented Feb 10, 2025

Uh oh!

neoneye commented Feb 10, 2025

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants