Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add commands/spideynolove/memory/extractgit clone --depth 1 https://github.com/spideynolove/memoryWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00015 | $0.01263 |
| Opus 5 | $0.00008 | $0.00632 |
| Sonnet 5 | $0.00003 | $0.00253 |
| Haiku 4.5 | $0.00002 | $0.00126 |
Grade A, and why
extract scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 151 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Multi-Dimensional Memory Extraction
Run three parallel analytical lenses on a diary entry or described event, synthesize with epistemic humility, then store the result via store_memory.
Input
The user provides one of:
- A diary entry path:
/extract .claude/diary/YYYY-MM-DD-session-N.md - A free-form description of an event:
/extract "I spent 20 minutes debugging X before realizing Y" - "last" to process the most recent diary entry automatically
If "last":
ls -t /home/hung/MCPs/memory/.claude/diary/*.md 2>/dev/null | head -1
Stage 1: Three Parallel Analyses
Run these three lenses mentally in parallel on the input:
Lens A — Root Cause (Five Whys)
Ask why five times to find the fundamental cause:
- What happened on the surface?
- Why did that happen? (immediate cause)
- Why did that happen? (contributing factor)
- Why did that happen? (systemic cause)
- Why did that happen? (root cause)
Extract: The deepest "why" — what fundamental pattern or assumption caused this?
Lens B — Psychological Driver (Hidden Motivation)
Uncover the real motivation behind the behavior:
- What was I trying to achieve?
- What insecurity or assumption might be driving this?
- What would never be admitted out loud?
- What need is being served by this pattern?
Extract: The hidden motivation that explains why this pattern persists.
Lens C — Prevention Strategy (Systems Thinking)
Identify feedback loops and build a gate function:
- What incentive structure caused this behavior?
- What feedback loop reinforced it?
- What corrective mechanism was missing?
- Gate function: What check would have prevented this?
Extract: A concrete prevention strategy — what to check before acting to avoid this mistake.
Stage 2: Epistemic Humility Check
Before concluding, verify:
- Confidence (1–5): How certain am I?
- Assumptions: What am I assuming is true?
- Alternatives: Could this be interpreted differently?
- Evidence: What would make me wrong?
- Ambiguity: Is this a genuine tradeoff where both approaches have merit?
- Lesson type: Is this about HOW we solved it (methodology) or WHAT the solution was (technical)?
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- yesterday First seen · 151 lines · 15 tokens per session scan A e277921ee204
extract is a command published in the GitHub repository spideynolove/memory (3 stars, last pushed 5mo ago), licensed MIT. It adds 15 tokens to every session and 1,263 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
learn
Force claude-smart to extract learnings from this session now.
component
Scaffold a new React component grounded in the paper-mono primitives. Requires explicit kind or a nearest-existing-component match. No empty divs, no speculative scaffolding.
memory-store
Store an insight, decision, or pattern to memory.
review
Cold re-quiz on code that already shipped — your own session commits, not the change in front of you.
no-vibe
Enter no-vibe mode in OpenCode (tutor mode, no direct project file writes).
teach-me-testing
Teach testing progressively through structured sessions. Use when user says ""lets learn testing"" or ""I want to study test practices"".