
Claude Skills by crescerelabs
github.com/crescerelabsExplains the available Sentience slash commands, what they do, and the boundaries (latest session only; no historical browsing).
Per-turn undeclared-intent drift drill-down for the latest Claude Code session — where the agent drifted from declared intent.
View the active Sentience governance profile — the rules currently gating this run. View-only; profile editing stays in the CLI.
One-command Sentience pulse for the latest Claude Code session. Composes undeclared-intent + policy-violation burn rate + advisory-flag summary; renders inline.
Retrospective review of your existing Claude Code history — which sessions wrote outside the project they were working in. Local and read-only; no prompt content is inspected.
Sentience capture status — is the Claude Code hook wired and capturing governance events? Quick health check.
Per-rule policy-violation drill-down for the latest Claude Code session — the compute burned per rule (POL-001..POL-005).