Rlm Gap Fill

RLM Distill Skill

Purpose

Distill (summarize) files directly into the RLM caches using the agent's own intelligence instead of a slow local Ollama model (like ibm granite or qwen2:7b on an M1 mac or windows machine). The agent reads the file, writes a summary, and updates the cache JSON in-place.

Should leverage frontier model ability to produce a superior summary for a file faster than can be done with a slow local CPU.

🎯 Agent Execution Protocol (The "Brain Upgrade")

Context: You (the Agent) are faster and smarter than the local distiller.py script. The local Ollama model (e.g., granite or qwen2) is often too slow on user hardware. Directive: When asked to "update RLM" or "summarize x", DO NOT run python distiller.py immediately.

Instead, you must act as the distiller yourself.

Protocol

1. NEVER RUN distiller.py: You are strictly forbidden from running python distiller.py or falling back to the local script, regardless of how many files are missing.
2. Act as the Intelligence Engine (Agent Distillation):
   • Call the rlm-distill pseudo-command (or just manually read) for each file.
   • Read the entire target file deeply (view_file).
   • Create a high-quality summary of the content yourself.
   • Execute inject_summary.py to manually inject your JSON/Text summary string into the cache structure.
   • Log: "Updated cache for [file] via Agent Distill."
3. Batching (The Swarm Protocol): If there are hundreds of files missing, DO NOT attempt to do them manually. You must use the parallel Multi-CLI Swarm engine:

   python3 plugins/agent-loops/skills/agent-swarm/scripts/swarm_run.py --job plugins/rlm-factory/resources/jobs/rlm_chronicle.job.md --dir <TARGET_DIR>
   

   This script (swarm_run.py) natively supports --engine [claude|gemini|copilot] and will dispatch the RLM summarization job in parallel.

Why This Exists (The RLM Philosophy)

The fundamental purpose of the recursive language model (RLM) cache is "Read Once, Cache Forever."

You should perform a deep, comprehensive read and summarize the file with an exceptionally good summary once. The goal is to entirely remove the need for you to read those complex files many times just to figure out what they do.

The workflow is:

1. You read the RLM cache summary (which you created once).
2. You immediately understand what the plugin/tool/document does without opening it.
3. If, and only if, the task requires deep code-level modification of that specific file, you trigger the "recursion" and read the full source file again.

The existing plugins/rlm-factory/skills/rlm-curator/scripts/distiller.py calls Ollama locally, which:

• Takes 3-5 minutes per file on M1 Mac hardware.
• Produces lower-quality summaries than a frontier model.
• Frequently fails ([DISTILLATION FAILED]).
• Requires Ollama to be running.

The agent (Claude, Gemini, Antigravity) is already a drastically better summarizer. This skill explicitly makes you the distillation engine.

The Two Caches (Example Configuration)

Projects configure their own caches in rlm_profiles.json. A common setup is:

First Run?

If no cache exists yet, use the rlm-init skill to interactively set up cache location, manifest, and .env config before distilling.

Cache Entry Schema

Summary Cache Entry (docs/markdown)

{
  "path/to/file.md": {
    "hash": "<content_hash_or_manual_marker>",
    "summary": "Plain text summary of the document...",
    "file_mtime": 1234567890.0,
    "summarized_at": "2026-02-11T18:30:00Z"
  }
}

Tool Cache Entry (code/scripts)

{
  "plugins/path/to/script.py": {
    "hash": "<content_hash_or_manual_marker>",
    "summary": "{\"purpose\": \"...\", \"layer\": \"...\", \"usage\": [...], \"args\": [...], \"inputs\": [...], \"outputs\": [...], \"dependencies\": [...], \"key_functions\": [...], \"consumed_by\": [...]}",
    "file_mtime": 1234567890.0,
    "summarized_at": "2026-02-11T18:30:00Z"
  }
}

Note: The tool cache summary field is a JSON string (not a nested object), matching the existing distiller's output format.

The Agent Gap-Fill Workflow (Procedure)

When the user asks you to "audit and update the RLM cache" or "distill missing files", follow this exact cyclical workflow.

1. Identify Gaps & Generate Task List

First, run the RLM Auditor to generate an actionable checklist of all files missing from the cache.

# Generate the full checklist of missing files for the target profile
python3 plugins/rlm-factory/skills/rlm-curator/scripts/inventory.py --profile <profile_name> --tasks

If the gap is massive (e.g., >20 files), you should pivot to --engine copilot (or gemini) and execute swarm_run.py against the generated checklist:

python3 plugins/agent-loops/skills/agent-swarm/scripts/swarm_run.py --engine copilot --files-from rlm_distill_tasks_<profile_name>.md --job plugins/rlm-factory/resources/jobs/rlm_chronicle.job.md

This generates rlm_distill_tasks_<profile_name>.md in the project root, containing checkboxes and pre-written inject_summary.py commands for every missing file.

2. Deep File Analysis

Open the rlm_distill_tasks_*.md file and begin working through the checklist. For each file:

1. Open and thoroughly read the source file (e.g. using view_file or cat).
2. Do not skip or skim. You must understand the file's purpose, architecture, inputs, and outputs.

3. Generate Summary & Exact Prompt Alignment

Before writing the summary, you MUST align your output exactly with the rigorous standards defined in the official RLM prompts.

For code/scripts (Tool Cache / tools profile): Read and strictly adhere to the JSON schema demanded in:

plugins/tool-inventory/resources/prompts/rlm/rlm_summarize_tool.md Your output must be the raw, stringified JSON object matching that exact schema.

For docs/markdown (Summary Cache / project profile): Read and strictly adhere to the high-fidelity architectural criteria demanded in:

plugins/rlm-factory/resources/prompts/rlm/rlm_summarize_general.md Your output must be a dense, signal-heavy text summary.

4. Inject Summary & Track Progress

Execute the pre-written inject_summary.py command found in your task list for that file, replacing the placeholder with your generated string:

python3 plugins/rlm-factory/skills/rlm-curator/scripts/inject_summary.py --profile <profile_name> --file "path/to/file.py" --summary "YOUR COMPLETE SUMMARY TEXT OR JSON"

1. After the injection succeeds, update the checkbox ([x]) in the rlm_distill_tasks_*.md file to track your progress.
2. Proceed to the next missing file on the list.

5. Verify & Re-Audit

Once you have completed the checklist (or if you are ending your session), run the auditor again without the --tasks flag to prove the gap has been closed:

python3 plugins/rlm-factory/skills/rlm-curator/scripts/inventory.py --profile <profile_name>

Quality Guidelines

Signal Over Noise

• Every summary should be Signal: a reader should learn the essential purpose and architecture from the summary alone
• Avoid Noise: don't pad with obvious observations, don't repeat the filename as the description
• A good summary lets the agent decide whether to read the full file without actually reading it

Conciseness

• Summary cache: Target 2-5 sentences for simple docs, up to a paragraph for complex protocols
• Tool cache: Keep purpose to 1-2 sentences. Let the structured fields carry the detail.

First Principles

• Summarize what the file actually does, not what it says it does
• If the file has a grand description but trivial implementation, note the gap
• Cross-reference related documents where architecturally significant

Incremental vs Full

• Incremental (preferred): Fix [DISTILLATION FAILED] entries and distill new files only
• Full: Only needed if the cache is severely stale or corrupted

Integration with Existing Distiller

This skill complements the Ollama-based distiller — it doesn't replace the script. The script is still useful for:

• Batch processing hundreds of files unattended
• CI/CD pipelines where no agent is available
• Content hash tracking for change detection

The agent distillation is better for:

• Fixing failed entries quickly
• Distilling complex documents that need frontier-model comprehension
• On-demand updates during active sessions
• M1 Mac environments where Ollama is too slow

Related

• plugins/rlm-factory/skills/rlm-curator/scripts/distiller.py — Original Ollama-based distiller
• plugins/rlm-factory/skills/rlm-curator/scripts/rlm_config.py — Configuration and cache utilities
• plugins/rlm-factory/skills/rlm-curator/scripts/query_cache.py — Search the cache
• plugins/rlm-factory/resources/rlm_manifest.json — Defines which directories get distilled