Skip to main content

Prompt Engineering Overview

Benchmarked against: Anthropic — Prompt Engineering Overview Architecture: CLAUDE.md + Rules + Hooks + Skills Scope: How SuperPortia configures agent behavior through prompts

In SuperPortia, "prompt engineering" is not about crafting individual prompts — it's about designing the entire behavioral configuration stack that shapes how agents think, decide, and act.


The configuration stack


Progressive Disclosure (PD)

The key principle: load the minimum at startup, expand on demand.

LayerWhatWhen loadedToken cost
L1: CLAUDE.mdIdentity, boot sequence, key pointersAlways~500
L2: RulesBehavioral standards, governanceAlways~2,000
L3: SkillsSpecific procedures, SOPsOn /command500-2,000 each

Why PD matters

Without PD, loading everything at startup would cost:

  • CLAUDE.md: ~500 tokens
  • All rules: ~2,000 tokens
  • All skills: ~5,600 tokens
  • Total: ~8,100 tokens — wasted on procedures that may never be used

With PD:

  • Baseline: ~2,500 tokens (CLAUDE.md + rules)
  • Per skill invocation: +500-2,000 tokens
  • Typical session: ~3,500-4,500 tokens

Prompt engineering in SuperPortia vs traditional

Traditional prompt engineeringSuperPortia prompt engineering
Craft one system promptDesign a layered configuration stack
Optimize for single taskOptimize for persistent agent behavior
Focus on output formatFocus on decision-making patterns
One-shot or few-shot examplesRules, skills, and precedent-based learning
StaticDynamic (hooks inject real-time context)

Core components

CLAUDE.md (L1)

The agent's identity card and boot instructions. Lean — just pointers, not procedures.

Contains:

  • Agent identity and role
  • Boot sequence (2 calls: heartbeat + mailbox)
  • Watch rule (timestamp every reply)
  • Pre-Flight Check reference
  • Knowledge retrieval flow
  • Key pointers to specs and rules

Does NOT contain:

  • Detailed procedures (those are in skills)
  • Full governance rules (those are in rules)
  • Historical decisions (those are in UB)

Rules (L2)

Behavioral standards that shape every action. Always loaded, always active.

RulePurpose
company-constitution.mdCore behavioral standards
cost-awareness.mdToken and engine discipline
agent-intelligence-protocol.md8 mandatory protocols
tech-freshness.mdPerishable knowledge defense
ub-governance.mdKnowledge management standards
gsta-alignment.mdStrategic alignment
compaction-recovery-protocol.mdMemory cliff defense
glossary.mdTerm navigation

Skills (L3)

On-demand SOPs loaded only when invoked with /command.

SkillCommandWhen to use
66s Review/66s-reviewMajor decisions, spec reviews
Create/createScaffolding new projects
Orchestrator/orchestratorMulti-agent scripts
Brain Lite/Mid/Pro/brain_lite etc.Engine mode switching
UI/uiFrontend component generation

Hooks

Event-driven context injection that fires at specific triggers:

HookTriggerWhat it injects
SessionStartNew sessionTimestamp, strategic anchor, intelligence
PreToolUseBefore tool callsCan gate operations
PostToolUseAfter tool callsCan log results

Design principles

PrincipleApplication
Lean over verboseCLAUDE.md has pointers, not content
Rules shape, skills guideRules define "always do this"; skills define "do this when asked"
Token-awareEvery token in the config stack costs money per conversation
Compaction-safeCritical facts must survive context compaction
Searchable knowledgeDetailed context lives in UB, not in prompts

PageRelationship
CLAUDE.md + Rules + HooksDetailed configuration guide
Prompt Caching (PD)Progressive Disclosure mechanics
Skills OverviewL3 skill reference
Context WindowsToken budget management