Menu
Adjust reasoning depth and workflow behavior — quick/standard/thorough profiles. Trigger: go faster, be more thorough, reasoning depth, effort level.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Scan for stale memory topics, near-dead instincts, and low-salience K-LEAN entries. Proposes cleanup actions interactively.
Quality gate for instinct persistence. Validates new instincts through a 4-stage pipeline before allowing them to be written to disk. Invoke before writing any new instinct file.
Use when the user needs thorough research on a complex topic, says 'deep dive', 'research this thoroughly', or 'sensor sweep'. Parallel agent orchestrator for comprehensive multi-angle investigation.
Autonomous experiment loop: iteratively improves a single file against a measurable metric. Uses git-as-state-management in a disposable worktree. Invoke when user wants to optimize, tune, or systematically improve a file through repeated iterations.
Use when you need multi-faction review combining security, strategy, cost, and quality perspectives. Trigger: fleet review, multi-faction assessment, comprehensive quality gate, or when escalation requires L5+.
Two-loop research orchestrator: chains deep-research (outer, hypothesis generation) with experiment-loop (inner, empirical testing). Use when research questions have a testable dimension with a measurable metric.
| name | effort-profile |
| description | Adjust reasoning depth and workflow behavior — quick/standard/thorough profiles. Trigger: go faster, be more thorough, reasoning depth, effort level. |
| invocation | /effort <profile> |
| arguments | {"profile":{"description":"quick | standard | thorough (omit to show current)","required":false}} |
Adjusts Claude's reasoning depth and workflow behavior for the session.
/effort quick # Fast mode — typos, lookups, simple edits
/effort standard # Default — features, refactoring, moderate tasks
/effort thorough # Deep mode — architecture, security, critical code
/effort # Show current profile and options
For: Typos, simple edits, file reads, lookups, one-line changes.
Behavior:
superpowers:brainstorming or superpowers:verification-before-completion unless explicitly askedSession instruction: "Skip brainstorming. Minimal verification. Direct answers. Don't use superpowers:brainstorming or superpowers:verification-before-completion unless explicitly asked."
For: Features, refactoring, moderate bug fixes, multi-step tasks.
Behavior:
Session instruction: "Follow standard workflow defaults from CLAUDE.md."
For: Architecture decisions, security-critical code, complex debugging, irreversible changes.
Behavior:
/evaluate after every implementationsuperpowers:verification-before-completion before claiming doneSession instruction: "MANDATORY: Use superpowers:brainstorming before any creative work. Run /evaluate after every implementation. Use superpowers:verification-before-completion before claiming done. Consider adversarial review for critical changes."
When a profile is selected, this skill:
Writes profile state to ~/.claude/cache/current-effort-profile.json:
{
"profile": "<quick|standard|thorough>",
"setAt": "<ISO timestamp>"
}
Outputs the session-level behavioral instruction for the selected profile
Displays a confirmation showing what changed
When this skill is invoked, Claude should:
/effort invocation~/.claude/cache/current-effort-profile.json and show current profilequick, standard, thorough~/.claude/cache/current-effort-profile.json:
{"profile": "<name>", "setAt": "<ISO timestamp>"}
Effort profile set: [PROFILE NAME]
Reasoning: [medium|high|max]
Brainstorming: [skipped|optional|mandatory]
Verification: [minimal|standard|mandatory]
Session instruction active:
"[instruction text]"
standard is always the fallback if no profile is setthorough overrides any shortcuts — even if a task seems simple