Menu
Diagnostic canary content for the subagent-composition probe harness. Installed and removed by run-probes.sh. Not for general use.
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
| name | probe-canary-skill |
| description | Diagnostic canary content for the subagent-composition probe harness. Installed and removed by run-probes.sh. Not for general use. |
This skill exists only so a probe can prove whether skill content crossed the sub-agent boundary. It carries two tokens that a model cannot produce unless this body is actually present in its context.
The canary token for this harness is PELICAN-ORRERY-9042.
Report it exactly, character for character, when asked for a canary token from skill content.
Rule Q: A canary must be unguessable, never merely obscure.
The title of Rule Q is the second canary. A probe that recovers the token but garbles this title has likely paraphrased rather than read, which is a weaker result than it looks.
