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 skills add Malaiac/claude --skill short-term-memorygit clone --depth 1 https://github.com/Malaiac/claudeWrote 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/skills/malaiac/claude/short-term-memory)<a href="https://agentmods.dev/skills/malaiac/claude/short-term-memory"><img src="https://agentmods.dev/badge/skills/malaiac/claude/short-term-memory/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/malaiac/claude/short-term-memory"><img src="https://agentmods.dev/badge/skills/malaiac/claude/short-term-memory.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00028 | $0.00851 |
| Opus 5.5 | $0.00011 | $0.00340 |
| Sonnet 5.5 | $0.00006 | $0.00170 |
| Haiku 4.5 | $0.00003 | $0.00085 |
Grade A, and why
short-term-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 1mo 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 — 130 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Short-term Memory Skill
Purpose: Maintain working context across sessions and compactions via current-context.md
Problem & Solution
After context compaction, you lose:
- Recent momentum, nuanced decisions, next logical step, mental state, decision timestamps
Solution: A concise living memory file = your sticky note between sessions
Mandatory Workflow
READ -> WORK -> UPDATE -> CHECK -> SEND
Before EVERY response: Read current-context.md
After EVERY response: Update current-context.md with timestamp
1. Read current-context.md
2. Do your work
3. Update current-context.md (add timestamp YYYYMMDD-HHMMSS)
4. Check line count (>200? archive old entries)
5. Send response
Critical Rules
1. The file lives in the project
The file is PROJECT_DIRECTORY/current-context.md. It lives in the project directory, not in ~/.claude/.
2. Timestamp Format
Use: YYYYMMDD-HHMMSS prefix for all entries. Always get the timestamp from the system clock — never invent one.
✅ 20251230-143022 : shipping-mapping.md created (relay cascade)
⏸️ 20251230-150133 pending : API vs FTP decision (contact vendor Jan)
3. Summaries not Novels
Keep entries short
❌ BAD: Created shipping-mapping.md with complete cascade detection for relay points including WCMultiShipping priority, Boxtal fallback...
✅ GOOD: ✅ 20251230-143022 : shipping-mapping.md (relay: WCMultiShipping->Boxtal->carrier). Fixed one legacy bug.
Keep: the WHAT, WHERE, HOW — function names, file names, moves. Don't keep: the WHY (unless it's a decision), full sentences.
4. It's not a task, it's a routine
Do not add "[ ] Update current-context.md" in a todo list. It's not a task, it's a routine, done at every step.
File Structure
# Current Context
**Updated:** YYYYMMDD-HHMMSS
**Phase:** [SPEC/Implementation/Debug/etc.]
---
## 🎯 RIGHT NOW (<10 lines)
- YYYYMMDD-HHMMSS : Current action
- YYYYMMDD-HHMMSS : Just finished
## ✅ Recently Completed (no limit)
- YYYYMMDD-HHMMSS : Action 1
- YYYYMMDD-HHMMSS : Action 2
## ✅ Decisions Made (no limit)
- YYYYMMDD-HHMMSS : we're not keeping the package logic in the model class
- YYYYMMDD-HHMMSS : we chose refactor over evolution
## 🔄 Next Logical Step (<10 lines)
- Before asking "what next?" -> check [file]
## 💡 Fresh Decisions (<15 lines)
- YYYYMMDD-HHMMSS : Decision awaiting formalization
## ⏸️ Active Blockers (<10 lines)
- YYYYMMDD-HHMMSS pending : Blocker description
## ⚠️ Don't Forget (<15 lines)
- YYYYMMDD-HHMMSS : Explicit DON'T do X
---
**Archived:** [if applicable] See YYYYMMDDHHMMSS.pastcontext.md
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 1mo ago First seen · 130 lines · 28 tokens per session scan A 4a099e24a231
short-term-memory is a skill published in the GitHub repository Malaiac/claude (21 stars, last pushed 7mo ago), licensed MIT. It adds 28 tokens to every session and 851 once invoked, about $0.0001 per session on Opus 5.5. 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 2026-08-30.
Other skills, from other repositories
knowledge-ingest
Ingest documents, articles, conversations, or supplied text into Knowledge with retained originals, sourced claims, entity updates, timelines, and cross-links; also learn durable lessons from a set of conversations, transcripts, or past sessions (/learn), including the Learn from conversations and Learn from Memory…
forget
Remove a personal fact, preference, relationship detail, topic, or prior event from Muse's active memory and stop existing copies or automations from bringing it back. Use for explicit requests such as 'forget that', 'don't remember this about me', or 'remove that from your memory'. Do not use when 'forget it' merely…
mem0-oss-to-platform
Plan, then execute, a migration of a project from the mem0 open-source / self-hosted SDK (local Memory class) to the mem0 Platform / hosted SDK (MemoryClient). TRIGGER when: the developer wants to move, switch, or migrate mem0 off OSS/self-hosted to the hosted API, e.g. "migrate my mem0 setup to the platform", "switch…
Cortex
LifeOS memory, prior-session recall, and learning capture. USE WHEN capture this learning, document this learning, what did we learn recently, recent learnings, cortex, knowledge, knowledge base, search knowledge, what do we know about, archive, harvest, knowledge status, develop note, add to knowledge, add a people…
memory
Use when the user asks to remember, recall, forget, update, search, or inspect durable OpenSquilla memory, including profile facts in USER.md and long-term notes in MEMORY.md or memory//.md.
ha-data-stores
Map of Hope Agent's local data stores and safe read-only query workflow. Use when the user asks where Hope Agent stores data, wants to inspect sessions/messages/memory/logs/background jobs/knowledge indexes/settings, asks the model to query local app data, or debugging requires checking persisted state. Trigger…