Back to skills
SKILL.md
Rlm Init
ASecurityInteractive RLM cache initialization for setting up a project semantic cache or adding a profile.
- 7 stars
- 0 votes
- 0 copies
- 0 views
- Added September 2, 2026
Security analysis
100/100Pro scans all 20 files and shows the line behind each finding
npx -y skills add richfrem/agent-plugins-skills --skill rlm-init --agent claude-codeAre you the author of Rlm Init?
Add the live security badge to your README. It updates with every re-scan.
[](https://www.skillsdirectory.com/skills/richfrem-rlm-init-agent-plugins-skills)---
name: rlm-init
plugin: agent-memory
description: Interactive RLM cache initialization for setting up a project semantic cache or adding a profile.
allowed-tools: Bash, Read, Write
---
# RLM Cache Initialization (`rlm-init`)
Initializes a new Recursive Language Model (RLM) semantic cache for zero-dependency high-speed memory.
## Contents
- [Critical Constraints](#critical-constraints)
- [Quick start](#quick-start)
- [Setup Modes](#setup-modes)
- [Workflow](#workflow)
- [Verification](#verification)
- [References](#references)
## Critical Constraints
1. **Zero-Dependency**: Uses standard library Python 3.8+ only; no external databases or daemons required.
2. **Profile Storage**: Profiles default to `.agent/learning/rlm_profiles.json`.
3. **Scripted Writes**: Never create cache markdown files manually; use `scripts/inject_summary.py`.
## Quick start
Check existing coverage or initialize a profile:
```bash
python3 scripts/inventory.py --profile project
```
## Setup Modes
- **Mode A (Standalone)**: RLM only. Fast keyword lookup across dense file summaries.
- **Mode B (RLM + vector-db)**: RLM keyword pre-filter into vector semantic search.
- **Mode C (RLM + obsidian-wiki-engine)**: RLM as wiki concept distiller.
- **Mode D (Super-RAG)**: Full stack combining RLM, vector store, and wiki concept nodes.
## Workflow
1. **Requirements Discovery**: Select target directories, file types, and cache destination.
2. **Profile Configuration**: Define profile entry in `.agent/learning/rlm_profiles.json`.
3. **Manifest Creation**: Define include/exclude patterns in `<name>_manifest.json`.
4. **Audit Gaps**: Run `python3 scripts/inventory.py --profile <name>` to list uncached files.
5. **Initial Distillation**: Deep read and inject summaries via `python3 scripts/inject_summary.py`.
6. **Report Readiness**: Confirm initialized profile paths and ready state to the user.
## Verification
Confirm 100% cache coverage with the inventory script:
```bash
python3 scripts/inventory.py --profile project
```
## References
- [acceptance-criteria.md](references/acceptance-criteria.md) — Acceptance criteria for RLM cache bootstrapping.
- [rlm-bootstrap-guide.md](references/rlm-bootstrap-guide.md) — Comprehensive bootstrap manual and mode details.
Files in this skill
- SKILL.md
- acceptance-criteria.md
- assets/resources/distiller_manifest.json
- assets/resources/manifest-index.json
- assets/resources/prompts/rlm/rlm_summarize_general.md
- assets/resources/prompts/rlm/rlm_summarize_tool.md
- assets/resources/rlm_manifest.json
- assets/rlm_profiles.json
- assets/rlm_summary_cache_manifest.json
- assets/rlm_tools_manifest.json
- evals/evals.json
- evals/results.tsv
- references/acceptance-criteria.md
- references/cheapest_models.json
- references/cheapest_models.md
- requirements.txt
- scripts/cleanup_cache.py
- scripts/distiller.py
- scripts/inject_summary.py
- scripts/inventory.py
Attribution
Comments
Loading comments…