skills.rest
Arasz
AraszCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

memory-quality-logging

Documents the removed file-based memory_search grading workflow and its server-side replacement plan.

2|Updated Aug 2, 2026
Search Tags:#deprecation#memory#telemetry#jsonl#grading#search quality#ai-raccoon
One-click install
npx skills add https://github.com/Arasz/ai-raccoon --skill memory-quality-logging-arasz
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: memory-quality-logging
Source: https://github.com/Arasz/ai-raccoon/tree/main/.ai-badger/skills/learned/uncategorized/memory-quality-logging
Command: npx skills add https://github.com/Arasz/ai-raccoon --skill memory-quality-logging-arasz

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve? This Skill prevents agents from running a deprecated memory-quality grading recipe by documenting that the file-based memory_search grading pipeline (AI_BADGER_MEMORY_GRADE, memory_grade.py, JSONL logs) was removed on 2026-08-11 and pointing to its replacement. ## Core Features & Use Cases - Deprecation Notice: Clearly states that the env var, grading script, and JSONL log writers no longer exist in ai-badger and ai-raccoon. - Replacement Pointer: Directs users to the server-side search-quality metric plan (docs/plans/2026-08-11-search-quality-metric-plan.md) with 100% search capture and follow-through measurement. - Interim Guidance: Until the replacement ships, grade search quality manually in-session and note rated searches in the session report. - Use Case: An agent about to run memory_grade.py grade <ts> <1-5> or scripts/audit_coverage.py reads this Skill first and learns those inputs are no longer written, avoiding wasted effort. ## Quick Start Read this Skill before attempting any memory_search quality grading to confirm the file-based logging workflow is removed and grade manually in-session instead.

Frequently Asked Questions about memory-quality-logging

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I grade memory_search quality after the logging removal?▼

Grade search quality manually in-session by noting rated searches in the session report, with no file logging. The planned replacement is a server-side quality table described in docs/plans/2026-08-11-search-quality-metric-plan.md with 100% search capture.

Why does memory_grade.py no longer exist?▼

The file-based grading workflow was removed on 2026-08-11 by owner decision (task mem-cleanup). The script was deleted from ai-badger in PR #373, and the ai-raccoon provider's memory-operations.jsonl writer was removed in PR #259.

Does AI_BADGER_MEMORY_GRADE still enable grading hooks?▼

No, the environment variable is now inert. Setting it has no effect because the hook script and log writers it controlled no longer exist in the plugin or framework.

Can I still run audit_coverage.py on old grading logs?▼

The script can still parse historical JSONL files that remain on disk, but its inputs are no longer written by any system. Running it produces only a static snapshot of pre-removal coverage, not current data.

What replaces the memory-quality JSONL logging pipeline?▼

A server-side quality table in the ai-raccoon repository, planned in docs/plans/2026-08-11-search-quality-metric-plan.md. It provides 100% search capture plus follow-through measurement without any JSONL files.

Related Skills

Explore More Software Engineering Skills→
nano-stepnano-step
3

nano-brain

Enable persistent memory and hybrid search across sessions, codebase, notes, and logs.

Official
Advanced
ZeptinyZeptiny

ce-optimize

Automate iterative optimization loops with parallel experiments and metric-driven evaluation.

Community
Advanced
DariusCorvusDariusCorvus
37

gstack

Automate browser-driven QA testing with a headless Playwright-based CLI.

Community
Advanced
ProfSynapseProfSynapse
26

dataset-improvement

Extract, review, and replace lines in JSONL dataset files with automatic backups.

Community
Intermediate