readiness-frame-modernization

star 1

Retrieves PWS-implied current methods and customer-grounded innovation openings; emits modernization_handoff.json with current_methods[] and innovation_opportunities[]. Use for readiness-frame-modernization solo/chain node or Mission Readiness Frame modernization slice. Do not use for pains, eval, workload, tea-leaves, or win-themes.

BdM-15 By BdM-15 schedule Updated 6/16/2026

name: readiness-frame-modernization description: > Retrieves PWS-implied current methods and customer-grounded innovation openings; emits modernization_handoff.json with current_methods[] and innovation_opportunities[]. Use for readiness-frame-modernization solo/chain node or Mission Readiness Frame modernization slice. Do not use for pains, eval, workload, tea-leaves, or win-themes. license: MIT metadata: personas_primary: capture_manager personas_secondary: [proposal_manager] shipley_phases: [capture, strategy] capability: analyze skill_role: slice skill_family: readiness-frame skill_family_label: Mission Readiness Frame runtime: tools category: capture_intelligence version: 2.0.0 status: active research_harness: plan_surfaces_path: references/plan_surfaces.json deliverables: [modernization_handoff.json] frame_artifact: modernization_handoff.json min_kg_chunks_passes: 2 max_turns: 12

depth_extension_turns: 0

Readiness Frame — Modernization / Innovation

Micro-skill for current methods and innovation opportunities only.

Read before writing:

  • references/readiness_output_contract.md
  • references/modernization_handoff_schema.md

Out of scope

Ignore chain prompts asking for pains, eval crosswalk, workload enablers, tea-leaves, or win-themes. Do not emit customer_pain_points[] or win_theme_candidates[] in this handoff.

Workflow

1. Inventory (one call)

kg_entities once with system, tool, requirement, performance_standard. Do not repeat.

2. Retrieve (one kg_chunks per surface — one per turn)

Follow artifacts/retrieval_plan.json sequentially. Surfaces: methods_modernization, then innovation_inquiry.

For each turn while plan_complete is false:

  • Read next_step.suggested_query
  • Run exactly one kg_chunks
  • Advance on the following turn

Never fire multiple kg_chunks in one assistant turn.

When every surface is retrieved or saturated, stop calling kg_chunks and kg_entities.

3. Draft (write handoff once)

When plan_complete: true, write artifacts/modernization_handoff.json per references/modernization_handoff_schema.md:

  • current_methods[] — ≥3 objects with method, implied_by, tooling, fit_to_scope, source_chunk_ids[] — PWS/QASP/CDRL-implied incumbent processes only
  • innovation_opportunities[] — ≥2 objects with opportunity, value (quality up / cost down / both), customer_grounded, fit_to_scope, source_chunk_ids[]
  • claim_gaps[] — honest named gaps

Use real source_chunk_ids from the scratchpad. Do not emit plain strings for method/innovation rows. If write_file is blocked in retrieve phase, run the next planned kg_chunks instead of retrying write.

Do not read_file the scratchpad — evidence is in tool results.

4. Stop

After handoff JSON is written, stop. Platform finalize runs outside this loop.

Retrieval discipline (latency)

Target: ≤12 turns, ≤120s on mcpp_rfp-class packages.

Step Budget
kg_entities 1 turn
kg_chunks (2 surfaces) 2 turns
write handoff 1–2 turns

Quality bar (production / platform gate)

Capture-grade handoff — not just non-empty JSON:

  • current_methods ≥ 3 object rows, each with method, implied_by (≥50 chars), source_chunk_ids
  • innovation_opportunities ≥ 2 object rows, each with opportunity, value (≥50 chars), source_chunk_ids
  • fit_to_scope = high | medium | low when set; no sibling-slice fields (pains, eval, win-themes)
  • claim_gaps[] names missing tooling/method evidence honestly when thin

Eval cases: evals/evals.json.

Install via CLI
npx skills add https://github.com/BdM-15/proj-theseus --skill readiness-frame-modernization
Repository Details
star Stars 1
call_split Forks 0
navigation Branch main
article Path SKILL.md
More from Creator