skillfed

decision-analysis

Decision Analysis exposes assumptions, failure paths, and success conditions for high-stakes design choices. It uses premortem and backcast reasoning to rank options against constraints, reversibility, and evidence quality, then returns a risk-calibrated recommendation. The skill routes obligations and rationale to their authoritative destinations—RFC, ADR, Work Item, or verification—rather than creating artifacts itself.

Decision Analysis stress-tests design options using premortem and backcast reasoning to deliver risk-calibrated recommendations.

AI-generated summary based on this skill's SKILL.md

239 12 MIT updated by govctl-org

Install

govctl-org/govctl/decision-analysis · repository language: Rust

git clone https://github.com/govctl-org/govctl
cp -r govctl/.claude/skills/decision-analysis ~/.claude/skills/decision-analysis
npx skillfed install govctl-org/govctl/decision-analysis

Frequently asked questions

AI-generated answers based on this skill's SKILL.md and metadata

How do I analyze design options before committing to a consequential choice?

Decision Analysis stress-tests your options using premortem and backcast reasoning. It exposes assumptions, failure paths, and success conditions for each alternative, then ranks them against constraints like reversibility, evidence quality, and governance fit. The skill returns a risk-calibrated recommendation aligned to your decision authority—whether that's an RFC, ADR, Work Item, or verification gate—without creating artifacts itself.

What could go wrong with this design option?

Decision Analysis surfaces failure paths by working backward from each option's worst plausible outcome (premortem) and forward from your success criteria (backcast). This exposes hidden assumptions about dependencies, capacity, and integration points. The skill identifies which failure modes are reversible, which carry high consequence, and what evidence would either validate or invalidate your choice before you lock it in.

How do I know if my design choice is reversible?

Decision Analysis evaluates reversibility as part of its option ranking. It assesses whether you can unwind each choice without cascading costs, and flags decisions that create lock-in or path dependency. Reversible choices carry lower risk and may justify faster decisions; irreversible ones warrant deeper analysis, more evidence, and higher governance scrutiny.

When should I do deeper analysis versus a quick comparison?

Decision Analysis determines analysis depth by weighing consequence, reversibility, and evidence quality. High-consequence, irreversible choices with weak evidence justify full premortem, backcast, and assumption validation. Lower-stakes or reversible decisions may warrant quick comparison only. The skill routes the rationale and next steps to the right decision authority so you know what depth is justified.

How does Decision Analysis compare alternatives before an ADR?

Decision Analysis ranks alternatives by stress-testing each against constraints, failure modes, and success conditions. It identifies which evidence would change the ranking or trigger a pivot, and calibrates risk to your governance context. Rather than writing the ADR itself, it routes the comparison, assumptions, and recommendation to your decision authority—enabling them to author a well-grounded record.

What assumptions and success conditions does Decision Analysis expose?

Decision Analysis uses premortem and backcast to surface assumptions about dependencies, capacity, team skill, vendor stability, and integration points. It maps success conditions—what must be true for each option to deliver value—and failure paths that would violate them. This exposure lets you validate assumptions early and identify which evidence matters most before committing.

SKILL.md

rendered from the published skill — quoted content, verbatim

Decision Analysis

Expose assumptions, failure paths, success conditions, and option-changing evidence before a consequential design choice is recorded.

This is a reference skill used by discuss and adr-writer. It analyzes; it does not create artifacts, mutate lifecycle state, write implementation, create Work Items, or perform VCS operations.

When Analysis Helps

Use deeper analysis when plausible options have non-obvious trade-offs, the choice is expensive to reverse, consequences cross important boundaries, or a wrong assumption could create normative, compatibility, data, security, or operational risk.

Use a compact comparison for reversible, well-understood choices. Skip this skill when an existing normative RFC already determines the answer or when the task is a routine implementation choice with no durable architectural consequence.

Hard Stops

  • Do not treat invented scenarios, unsupported probabilities, or model confidence as evidence.
  • Do not

(truncated - see the full file via the links below)

Read as markdown · JSON record · Browse the source repository

File tree — 1 file
.claude/skills/decision-analysis/SKILL.md

Related skills

Tags

premortem-analysis risk-calibration assumption-validation option-comparison failure-path-mapping backcast-reasoning reversibility-assessment governance-alignment trade-off-evaluation uncertainty-reduction