field-dev-code-recon

Extract field development information from external sources (LinkedIn posts, technical content), map against digitalmodel codebase coverage, document gaps, and create actionable GitHub issues.

SKILL.md

GitHub

Field Development Code Reconnaissance

Trigger

User asks to review field development content from external sources (LinkedIn posts, articles, documentation) and extract actionable intelligence with codebase mapping.

Workflow

Phase 1: Content Extraction

  1. Use browser_navigate to load the source URL
  2. Dismiss sign-in dialogs (click @e1 button if present)
  3. Use browser_snapshot with full=true to capture complete content
  4. Extract: component names, standards references, operational workflows, hazards, mitigations
  5. Close browser when done

Phase 2: Code Coverage Mapping

Map extracted components against digitalmodel using STRONG/PARTIAL/NO coverage tiers:

  • STRONG: Native Python module with tests covering the functionality
  • PARTIAL: OrcaFlex skill only, incomplete module, or partial standards coverage
  • NO COVERAGE (GAP): No code exists at all

Use this terminal command to survey existing modules:

cd /mnt/local-analysis/workspace-hub/digitalmodel
find src/digitalmodel -name '*.py' -not -path '*/__pycache__/*' | sort

For standards/skills coverage:

skills_list  # in Python/hermes context

Mapping template:

Component Standard/Ref Coverage digitalmodel Module
... API 17D STRONG subsea/trees/
... DNV-RP-F101 PARTIAL (skill only)
... -- GAP --

Phase 3: Documentation

Create docs/field-development/<topic-name>-mapping.md with:

  • Source descriptions and extraction date
  • Component inventory with standards references
  • Operational workflow diagrams (ASCII)
  • Full coverage map (STRONG/PARTIAL/GAP tables)
  • API/standards quick reference table
  • Architecture diagram (ASCII) showing component relationships

Phase 4: Issue Creation

Create one GitHub issue per GAP or significant PARTIAL component:

cd /mnt/local-analysis/workspace-hub/digitalmodel
gh issue create \
  --title "Implement <component name> (<standard ref>)" \
  --label "cat:engineering,enhancement,priority:medium" \
  --body "<structured body>"

Issue body must include:

  • Background/context linking to the mapping doc
  • Problem statement
  • Scope (module path, key capabilities)
  • Acceptance criteria (checklist of files to create)
  • Standards references
  • Related issues

Phase 5: Commit & Push

git add docs/field-development/
git commit -m "docs: <topic> coverage mapping (#issue-numbers)"
git push

Pitfalls

  • search_files with very broad paths (entire workspace) hits output limits — prefer targeted terminal find commands in digitalmodel/src/
  • LinkedIn content is gated behind sign-in — dismiss the dialog, then snapshot
  • Don NOT create issues for STRONG coverage areas
  • Prioritize medium over low for core subsea components, low for operational/installation topics
  • All issues should reference the mapping doc at top of body
  • Commit documentation BEFORE creating issues (or reference pending issue numbers)

Standards Quick Reference

API 17 subsea: 17A=General, 17B=Flexible pipe, 17D=Trees, 17E=Umbilicals, 17F=Controls, 17G=Workover riser, 17H=ROV interface, 17P=Manifolds/Structures, 17R=Connectors/Jumpers, 17W=Capping stack

Key DNV standards: DNV-OS-F101 (pipe), DNV-RP-F105 (free span), DNV-RP-F109 (on-bottom), DNV-RP-B401 (CP), DNV-RP-H103 (hydro), DNV-ST-N001 (marine ops)