Stats
Actions
Tags
From power-trio
Centralized configuration for the Power Trio workflow. Reference this file for consistent settings. These values must be adhered to.
How this skill is triggered — by the user, by Claude, or both
Slash command
/power-trio:parametersThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Centralized configuration for the Power Trio workflow. Reference this file for consistent settings. These values must be adhered to.
Centralized configuration for the Power Trio workflow. Reference this file for consistent settings. These values must be adhered to.
| Parameter | Value | Description |
|---|---|---|
MAX_PLAN_ITERATIONS | 3 | Maximum rounds of plan revision before escalating to user (expected convergence: 1-3 iters) |
MAX_BLOCKS_PER_AGENT | 3 | Blocks before escalating concern to documented risk |
MAX_OPEN_QUESTIONS | 2 | Maximum accepted open questions at completion (prefer 0, accept 1-2 if documented as risk) |
| Parameter | Value | Description |
|---|---|---|
MAX_TEST_ITERATIONS | 2 | Maximum rounds of test revision |
MAX_IMPL_ITERATIONS | 3 | Maximum rounds of implementation revision |
MAX_TEST_FIX_ATTEMPTS | 2 | Maximum attempts to fix failing tests |
MAX_BLOCKS_PER_AGENT | 3 | Blocks before escalating concern to documented risk |
EXPECTED_TEST_CONVERGENCE | 2 | Expected maximum test iterations for normal cases |
EXPECTED_IMPL_CONVERGENCE | 3 | Expected maximum implementation iterations for normal cases |
| Parameter | Value | Description |
|---|---|---|
MAX_PHASE2_LOOPS | 2 | Maximum Phase 2↔3 loops before escalating to user |
| Parameter | Value | Description |
|---|---|---|
FILES_CHANGED_THRESHOLD | 5 | Trigger Phase 3 if files changed ≥ this |
LOC_CHANGED_THRESHOLD | 300 | Trigger Phase 3 if LOC changed ≥ this |
| Parameter | Value | Description |
|---|---|---|
RISK_NONE | "NONE" | No risk detected - skip specialist |
RISK_LOW | "LOW" | Low risk - run specialist with minimal scope |
RISK_MEDIUM | "MEDIUM" | Medium risk - run specialist with standard scope |
RISK_HIGH | "HIGH" | High risk - run specialist with full scope |
| Parameter | Value | Description |
|---|---|---|
MAX_FILES_PER_SPECIALIST | 5 | Maximum files to pass to each specialist |
| Severity | Max Count | Description |
|---|---|---|
MAX_CRITICAL_FINDINGS | 2 | Maximum CRITICAL findings per specialist |
MAX_HIGH_FINDINGS | 1 | Maximum HIGH findings per specialist |
MAX_MEDIUM_FINDINGS | 1 | Maximum MEDIUM findings per specialist |
MAX_LOW_FINDINGS | 1 | Maximum LOW findings per specialist |
| Condition | Model | Description |
|---|---|---|
| Default | sonnet | Standard security review model |
| High-risk areas | opus | Escalate to opus for sensitive areas |
npx claudepluginhub thesimonho/artificial-jellybeans --plugin power-trioMaximizes test coverage for safety, then methodically increases build time, test time, complexity, LOC, and degrades performance
Orchestrates parallel agent teams for implementation, simplify/harden audits, and iterative fixes until code compiles cleanly, tests pass, and zero issues found. For multi-file features, hardening, or batch fixes.
Multi-phase code review pipeline with mechanical checks, graph-scoped context, parallel review agents, cross-agent deduplication, and structured output. Use for completed work reviews.