Skip to content

rfe-feasibility-review

Internal sub-agent (model: opus) launched per RFE by rfe.review. Acts as a senior engineer looking for blockers and risks rather than confirming the work is good. Reads the RFE task file (and Jira comment history if present), grounds its assessment in architecture context (LATEST_VERSION / PLATFORM.md plus active overlays filtered by status, release, and affected components), and reads any prior review report for re-reviews. Emits one of three verdicts -- feasible (can be built), infeasible (platform architecture fundamentally conflicts), or indeterminate (RFE too ambiguous to assess) -- distinguishing "capability doesn't exist yet" (feasible) from "architecturally incompatible" (infeasible). Named components absent from the architecture inventory are treated as strategy considerations, not blockers. Writes to artifacts/rfe-reviews/{ID}-feasibility.md.

Plugin: rfe-creator | Internal

Contract

Skill Contract canonical-skill-v1

Review a single RFE for technical feasibility, blockers, and alignment with technical strategy, and write the verdict.

Identity

Functions
review
Success
  • Produces a feasibility verdict (feasible/infeasible/indeterminate) for the RFE with blockers and a scope assessment.
  • Writes artifacts/rfe-reviews/{ID}-feasibility.md, grounded in architecture context when available.

Optimization Targets

task_success judge SKILL.md @ c8a2a1e

Invariants

Must Preserve
  • Do not block an RFE for a capability that simply does not exist yet — that is feasible, not infeasible.
  • Do not suggest implementation approaches — assess feasibility only.
Fixed Context
toolsRead, Write, Grep, Glob, Bash
knowledgerepository_contentpublic, task_inputtask_private

Traceability

Skill
.claude/skills/rfe-feasibility-review/SKILL.md

Diagram

rfe-feasibility-review diagram

Usage

/rfe-feasibility-review