shawnla90/recursive-drift

a methodology for building AI agents that develop memory, voice, and personality over time. CLI-first. zero API keys. built for Claude Code.

17Stars on the repository
4Mods indexed here, across every type
7mo agoLast push, which is what freshness is scored on
MITLicence, which decides whether bodies are shown

skill-name

01

shawnla90/recursive-drift

Skill Claude CodeCodex

One-line description of what this skill does and when it triggers.

not rated 17 7mo ago A 16 tokens original MIT

agent-routing

02

shawnla90/recursive-drift

Skill Claude CodeCodex

Decides whether a task needs Agent Teams, subagents, or a single focused session. Consult this skill before creating any team or spawning more than 2 subagents.

not rated 17 7mo ago A 37 tokens original MIT

context-handoff

03

shawnla90/recursive-drift

Skill Claude CodeCodex

Generate a context handoff document at the end of a session so the next session can pick up where this one left off. Auto-invoked when ending significant work, or manually via /handoff.

not rated 17 7mo ago A 45 tokens original MIT

discuss

04

shawnla90/recursive-drift

Skill Claude CodeCodex

Decision-lock workflow for complex features. Asks targeted domain-specific questions, locks scope, and writes a CONTEXT.md that downstream skills consume. Auto-invoked when a feature touches 3+ files across 2+ concerns, or manually via /discuss.

not rated 17 7mo ago A 56 tokens original MIT

At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: