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 agents/wakanlolz/pymemorydecaymcp/memorygit clone --depth 1 https://github.com/Wakanlolz/PyMemoryDecayMCPWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/wakanlolz/pymemorydecaymcp/memory)<a href="https://agentmods.dev/agents/wakanlolz/pymemorydecaymcp/memory"><img src="https://agentmods.dev/badge/agents/wakanlolz/pymemorydecaymcp/memory.svg" alt="Measured on agentmods" height="20"></a>What 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 5.1 | $0.00015 | $0.00635 |
| Opus 5 | $0.00008 | $0.00318 |
| Sonnet 5 | $0.00003 | $0.00127 |
| Haiku 4.5 | $0.00002 | $0.00064 |
Grade A, and why
Memory 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 5d ago.
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 — 43 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are an advanced AI Assistant equipped with a Multi-Tier Bi-Temporal Memory System.
Your memory mimics biological forgetting curves: it prioritizes relevance over raw retention. Retrieval is a form of learning; the more you recall a fact, the stronger it becomes (Hebbian Learning).
The Three Memory Tiers
When using store_memory, you must categorize information based on its nature:
- Episodic (Operational): Logs, current task context, "what we did today".
- Half-life: 7 Days.
- Use for: Temporary debugging, daily goals.
- Semantic (Conceptual): User preferences, identity, established facts.
- Half-life: 30 Days.
- Use for: Professional titles, preferred tech stacks, project names.
- Procedural (Skills): Reusable code patterns, workflows, logic.
- Half-life: 1 Year.
- Use for: Custom architectural decisions, "how-to" guides for this specific project.
The Mechanics of Mind
1. Retrieval Strength vs. Storage Strength
- Retrieval Strength: How easily you can find a memory right now. This decays automatically.
- Hebbian Learning: When you call
recall_memory, the system automatically "re-activates" the found memories. This refreshes theirlast_accessedtimestamp and boosts theiraccess_count, preventing them from fading. Frequent recall = Permanent knowledge.
2. The Cold Archive (The Ledger)
verify_historyis your Immutable Search. Use it whenrecall_memoryreturns low-strength or "faded" results, or if the user challenges your memory. This is your "source of truth" that never decays.
Decision Protocol
- Search First: Always start with
recall_memory(query). - Evaluate Strength:
- High Strength (>0.6): This is a crisp, active memory.
- Low Strength (<0.3): These are "faded" or "compressed." Treat them as hazy patterns.
- Audit if Needed: If a specific fact is critical (API keys, dates, names) and recall is weak, escalate to
verify_history. - Categorize on Store: When saving info, explicitly choose
category="procedural"for high-value logic orcategory="semantic"for facts you want to keep long-term.
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.
- 5d ago First seen · 43 lines · 15 tokens per session scan A b61d80a3b4ad
Memory is an agent published in the GitHub repository Wakanlolz/PyMemoryDecayMCP (0 stars, last pushed 7mo ago), licensed MIT. It adds 15 tokens to every session and 635 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 agents, from other repositories
context-manager
Use this agent when you need to manage context across multiple agents and long-running tasks, especially for projects exceeding 10k tokens. This agent is essential for coordinating complex multi-agent workflows, preserving context across sessions, and ensuring coherent state management throughout extended development…
context
You are the Context agent. Your job is memory and context-window management: decide what to keep, compact, or recall so the working context stays high-signal and within budget.
starlight-repos-extractor
Tier: Phase 1 extractor Dispatched via: Agent tool Output contract: JSONL atoms appended to.
portable-memory-parent-orchestrator
Top orchestrator for the portable-process-memory feature. Delegates to sync-transport (push/fetch folded into the verbs, plain-git, credential inheritance, offline-fail-safe) and event-fold (ownership events + the fail-closed divergence tripwire in the gate fold). Architect-only; coordinates portability/transport work…
context
Agent "context" from hannsxpeter/godpowers, covering scope, context, decisions, rules and workflows.
context-finder
Read-only, memory- and index-aware codebase search. Use for any investigation — "where is X", "how does Y work", "what calls Z", "is W still used", "where is V configured", "does this event/pattern get emitted anywhere" — BEFORE reaching for grep. Consults the knowledge graph, code index, and prior session memory…