Produce a rich, interactive HTML explanation of **how a feature/subsystem works and how the loops collectively built it**. The default unit of explanation is the **feature** (campaign/subsystem — e.g. token-telemetry, fleet-width, queue-integrity), aggregating ALL its contributing cycles across ba

A focused Go-test-engineering review. Run it over the changed `*_test.go` files (and any test harness / fixtures / test tooling) **after** writing tests and before merge. It does not replace `code-reviewer` — it adds the test-specific lens that general review misses.

> Interactive onboarding. Everything deterministic — detection, the per-phase model **recommendation**, the policy write, and verification — runs in the Go binary (`evolve setup detect|recommend|apply`). The only judgment left HERE, in your session (zero extra API cost), is **teaching the pipeli

DIFF_STAT=$(git diff HEAD --stat 2>/dev/null || echo "") if [ -n "$DIFF_STAT" ]; then REF="HEAD" # pre-commit: review uncommitted Builder changes else REF="HEAD~1" # standalone: review last committed change fi DIFF=$(git diff "$REF" --stat) CHANGED_FILES=$(git diff "$REF" --name-only) DIFF_FULL=

Use when authoring evals/predicates (Scout, TDD-Engineer), framing an audit (Auditor), or justifying which phases run (PhaseAdvisor/router). Codifies Google's 4-phase adversarial-testing methodology mapped onto evolve-loop's eval, audit, routing, and red-team surfaces.

Use when reviewing code changes for quality, security, performance, or maintainability issues and identifying simplification opportunities in a single token-efficient pass

> Why this skill exists: two generic review agents (go-reviewer + code-simplifier) > spent 197k + 136k tokens and 97 + 58 tool calls reviewing ONE staged diff > (2026-07-27). Measured waste: re-deriving repo context the caller already had, > re-running test suites the caller had already run, unbound

> The craft companion to `fable` (which governs how you *operate*; this governs the code you *produce*). Content is distilled from 2025-26 measured evidence — mutation-testing studies, million-commit analyses of AI-written code, and benchmarked skill collections — not textbook restatement. Sourc

> Fable 5's edge over other models is only partly raw capability; a large share is *process*: verify before building, attack your own premises, refuse symptom patches, parallelize ruthlessly, report your own failures unprompted. This skill encodes those as mechanically followable rules for ANY agent

This file exists solely to prove the flagreaders guard scans non-Go surfaces.

> Standing discipline for every coding session: write only what the task needs. The code ends up > small because it is necessary, not golfed. Default intensity: **full**. > Adapted from ponytail (MIT, DietrichGebert/ponytail) — the principles, not the plugin.

You turn a phase *idea* into a registered, advisor-selectable pipeline phase. All validation, collision-checking, scaffolding, and index rebuilding live in `evolve phases create` — your job is to design a good spec, call the command, and self-correct from its JSON envelope. Never write `.evolve/ph

Use when the user invokes /evo:verify-release or asks to check whether a release has propagated, whether the marketplace is up to date, or whether installed plugins reflect the latest version. Wraps legacy/scripts/release/marketplace-poll.sh for standalone post-publish verification.