UNPKG

aiwg

Version:

Deployment tool and support utility for AI context. Copies agents, skills, commands, rules, and behaviors into the paths each AI platform reads (Claude Code, Codex, Copilot, Cursor, Warp, OpenClaw, and 6 more) so one source of truth works across 10 platfo

91 lines (60 loc) 1.77 kB
# External Agent Loop - Continuation (Iteration {{iteration}}/{{maxIterations}}) ## Task Objective {{objective}} ## Completion Criteria {{completionCriteria}} --- ## Previous Session Summary **Status**: {{previousStatus}} ### What Happened {{previousOutput}} ### Last Analysis {{lastAnalysis}} --- ## Accumulated Learnings {{learnings}} ## Files Modified So Far - {{filesModified}} --- ## Instructions Continue working toward the objective. This is a **continuation** of previous work. ### First Steps 1. **Check internal Ralph state**: ``` /ralph-status ``` If active, resume with `/ralph-resume`. 2. **Check matric-memory for context**: ``` matric-memory get "ralph:external:{{loopId}}:*" ``` 3. **Review git history** for recent progress: ``` git log --oneline -10 ``` ### Continue Implementation If internal Ralph is not active, start a new internal loop: ``` /ralph "{{objective}}" --completion "{{completionCriteria}}" --max-iterations 10 ``` ### Apply Learnings Use the accumulated learnings above to avoid repeating mistakes and build on previous progress. ### Completion Signaling When complete, output: ```json {"ralph_external_completion": true, "success": true, "reason": "Task completed", "iterations": N} ``` If blocked, output: ```json {"ralph_external_completion": true, "success": false, "reason": "explanation", "iterations": N} ``` ## Session Context | Property | Value | |----------|-------| | External Iteration | {{iteration}} of {{maxIterations}} | | Loop ID | {{loopId}} | | Session ID | {{sessionId}} | | Time Remaining | {{timeRemaining}} minutes | | Budget Remaining | ${{budgetRemaining}} | ## Begin Continuation Review the previous session state and continue from where it left off.