Skip to content

title: "Briefing — "What should we investigate?"" source: "tasks/TFW-47__codex_adapter_shortcut_skills/research/iter1/1_briefing.md"

Briefing — "What should we investigate?"

Mindset: Strategist. You're planning an investigation, not doing it. Frame what matters. Resist solving. Test: "Can I explain WHY we're investigating this and what would change our approach?" Parent: HL-TFW-47 Goal: Every completed task produces a mandatory evidence/ folder with a structured template file capturing real verification results — closing the gap where evidence was defined but never physically materialized (0/38 tasks).

Research Plan

Gather (Explorer): - Survey all existing TFW artifact naming conventions (HL, TS, RF, RES, ONB, REVIEW, EV?) — identify naming patterns - Examine 3-5 completed task folders to catalog what was actually verified and how (inline RF §5 claims) - Identify candidate dimensions: naming scheme, template structure depth, proportionality mechanism, mandatory vs optional sections - Look at external evidence/verification template patterns (audit trails, compliance docs, test reports)

Extract (Analyst): - Build configuration space: name options × structure depth × proportionality approach - Cross-reference each combination against TFW values (trace-first, honest, structured, reproducible)

Challenge (Critic): - Test top candidates against 3 real past tasks (trivial fix, medium refactor, complex multi-phase) - Check: does the template capture what was verified without creating disproportionate overhead?

Hypotheses (from HL §10)

# Hypothesis HL Status
H1 A single structured evidence template file per task (not per-AC) is sufficient for traceability open
H2 The evidence template should include environment metadata (OS, tool versions, timestamps) for reproducibility open

Scope Intent

  • In scope: Evidence template file naming, internal structure, proportionality mechanism, alignment with TFW naming/values. How the template handles trivial vs complex tasks.
  • Out of scope: Codex adapter mechanics (iteration 2). Workflow integration details (HL Phase A deliverables 2-7, not research scope). How review workflow audits evidence (already defined in D52).

Guiding Questions

  1. Should the evidence file use a 2-letter abbreviation (like HL, TS, RF) or a longer name (like evidence.md)? Any preference?
  2. Should the template enforce per-AC evidence rows (one row per acceptance criterion) or allow freeform?
  3. Is there a minimum evidence bar even for trivial tasks (e.g., "I ran the linter, output was clean")?

User Direction

  1. Naming: EV abbreviation is accepted. Pattern: EV__TFW-{N}__{title}.md — consistent with HL/TS/RF/RES/ONB/REVIEW.
  2. Per-AC vs freeform: User doesn't know — needs research to decide.
  3. Minimum evidence bar: Needs empirical research from real TFW projects. User's most active projects: helpdesk, afd, tfw (this repo). Research should examine what was actually verified in completed tasks across these projects.

Stage complete: YES