Agent memory

Local consolidation for image-scoring-gallery — scripts and canonical store live in sibling image-scoring-pipeline (no duplicate dream engine in gallery).

When to use

• Session start: Read sibling ../image-scoring-pipeline/.agent-memory/memory.md; prefer repo evidence on conflict.
• Session end / milestone: Log durable learnings with /log-session.
• Periodic: Run /dream-memory, review changelog, /promote-memory after human approval.

Memory candidates

Use text|category|confidence on CLI or YAML memory_candidates:

Category	Use for
stable_fact	Stack, architecture, env
user_preference	Style, workflow, review prefs
working_rule	How to run tests, what not to touch
recurring_issue	Flakes, traps, env pain
successful_pattern	Approaches that worked
open_question	Unverified assumptions
deprecated	Superseded guidance

Confidence: low, medium, high.

Commands (gallery repo root)

python ../image-scoring-pipeline/scripts/agent-memory/log_session.py --summary "..." --outcome "..." --candidate "text|working_rule|high"
python ../image-scoring-pipeline/scripts/agent-memory/dream.py
python ../image-scoring-pipeline/scripts/agent-memory/promote_dream.py --dream ../image-scoring-pipeline/.agent-memory/dreams/<timestamp>.md
python ../image-scoring-pipeline/scripts/agent-memory/context.py

Transcript import: backend /import-transcripts and AGENT_MEMORY.md.

Dream review checklist

1. Open backend dreams/*-changelog.md — scan Uncertain / needs review.
2. Diff proposed dreams/*.md vs memory.md.
3. Redact anything sensitive; reject promotion if secrets might be present.
4. Promote only when the proposal is accurate and concise.

Safety

• Scripts block common secret patterns on write.
• Do not log secrets.json, .env, tokens, or personal library paths.
• Never edit memory.md directly during implementation work.

Reference

• .agent-memory/CURSOR_USAGE.md
• Backend .agent-memory/CURSOR_USAGE.md
• Backend docs/technical/AGENT_MEMORY.md