Boundary Probing

SkillSearch

Map parameter space, generate extreme values, test at boundaries, detect

Available today. Use it from your connected AI after setup.

Connect ahel once, and every AI you use reads what you have installed.

Then ask your AI: use the Boundary Probing skill

What this skill tells your AI

The instructions your AI receives, as published by yogsoth-ai/de-anthropocentric-research-engine in skills/boundary-probing/SKILL.md and read by ahel’s review.

Orchestration Steps

  1. Receive claim and parameter context from strategy
  2. Dispatch parameter-space-mapping to identify all dimensions
  3. For each dimension, dispatch extreme-value-generation:
    • Minimum, maximum, zero, negative, overflow
    • Type boundaries (empty, null, singleton, infinite)
  4. Dispatch breakpoint-detection at each extreme value
  5. Record pass/fail for each probe point
  6. Dispatch validity-envelope-construction to synthesize boundaries

Subagents

  • parameter-space-mapping
  • extreme-value-generation
  • breakpoint-detection
  • validity-envelope-construction

Termination Conditions

  • All dimensions probed at all boundary values (complete)
  • Breakpoint found in critical dimension (early report)
  • Budget exhausted (report partial envelope)
  • Dimension count exceeds budget (prioritize by sensitivity)

Available SOPs

Optional, no fixed order; the final leaf is always a sop.

SOPWhen to use
breakpoint-detectionTest a claim at extreme parameter values and detect the precise point where it breaks down.
extreme-value-generationGenerate boundary and extreme test values for a given parameter dimension to stress-test claims.
parameter-space-mappingIdentify all parameter dimensions along which a claim's validity might vary.
stress-test-validity-envelope-constructionSynthesize breakpoints across dimensions into a coherent validity envelope for a claim.

Signals

GitHub stars
469
Forks
37
Last commit
Sep 2026
Advanced
Catalog kind
skill
Gateway key
boundary-probing
Source
github.com/yogsoth-ai/de-anthropocentric-research-engine