How tropicron per-job memory works
Scanned 9/8/2026
Install to Claude Code
npx -y skills add pforret/TropicClaw --skill tropicron-memory --agent claude-codeInstalls into .claude/skills of the current project.
Are you the author of Tropicron Memory?
Add the live security badge to your README — it updates automatically with every re-scan.
[](https://www.skillsdirectory.com/skills/pforret-tropicron-memory-tropicclaw)More formats (shields.io, HTML) on the badges page.
---
name: tropicron-memory
description: How tropicron per-job memory works
user-invocable: false
---
## Per-Job Memory
Each tropicron job can have a `<job>.memory.md` sibling file for persistent context.
- When `memory: true` in frontmatter, the memory file is prepended to the prompt
- Claude is instructed to update the memory file with important findings
- Memory files are gitignored (runtime state, not config)
- Keep memory files concise (<100 lines) — summarize, don't accumulate
### Memory file location
`<job>.memory.md` lives next to `<job>.md` in the jobs directory.
### What to store in memory
- Last-seen values, counters, dates
- Recurring patterns discovered across runs
- User preferences learned from feedback
- Errors to avoid on future runs
- State that should persist (e.g., "last email checked: 2026-02-25")
Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.
No comments yet. Be the first to comment!