How do I install the Using Basemem skill?

Run npx skillmds@latest add ashvin-to/using-basemem in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Using Basemem safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Using Basemem?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Using Basemem free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Using Basemem?

ashvin-to (@ashvin-to) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Using Basemem by ashvin-to — Agent Skill

Name: Using Basemem
Author: ashvin-to

File contents

BaseMem — Memory + Code Intelligence Protocol

BaseMem provides persistent cross-session memory (planets, notes, decisions) and zero-RAM code intelligence.

Quick Decision Matrix

Task	Preferred Tool Call
Recall session / topic context	`getContext(topic="repo-name")`
Fast code search (line-level)	`code_find(query="...", grep=True)`
Read source code window	`code_read(path="...", offset=1, limit=50)`
Explore code structure	`code_explore(path="...")`
Code review blast-radius	`get_review_context(files=["a.py"])`
Log key decision / architectural fact	`logInteraction(topic="...", decision="...")`
Create / list task items	`task_create(...)` / `task_list(...)`
End session & save summary	`logInteraction(topic="...", summary="...", activity="done")`

Memory & Context Workflow

Session Start: Context is auto-injected. Skip getContext unless switching topics or refreshing context.
Logging Decisions: Keep facts $\le 30$ words; record essential architectural decisions immediately via logInteraction(topic, decision="...").
Session End: Call logInteraction(topic="...", summary="...", activity="done") when finishing work.

Code Intelligence Rules

Prefer code_find over code_read: Returns line numbers only (~0 extra context tokens).
Tight Reads: Keep code_read window sizes small (limit <= 50 lines).
Initial Indexing: If code_find returns empty results, run code_init(repo_path="...") first.

Gotchas

Avoid full-file reads unless strictly necessary — use code_find and code_read line slicing.
Always pass the exact topic (repository folder name) when logging decisions.
Keep task_create descriptions to single concise sentences ($\le 20$ words).

---name: using-basememdescription: BaseMem memory + code intelligence protocol. Triggers on: recall context, log decision, session lifecycle, code search, review blast radius, task management, planet memories, zero-RAM code graph.---# BaseMem — Memory + Code Intelligence ProtocolBaseMem provides persistent cross-session memory (planets, notes, decisions) and zero-RAM code intelligence.## Quick Decision Matrix| Task | Preferred Tool Call || :--- | :--- || Recall session / topic context | `getContext(topic="repo-name")` || Fast code search (line-level) | `code_find(query="...", grep=True)` || Read source code window | `code_read(path="...", offset=1, limit=50)` || Explore code structure | `code_explore(path="...")` || Code review blast-radius | `get_review_context(files=["a.py"])` || Log key decision / architectural fact | `logInteraction(topic="...", decision="...")` || Create / list task items | `task_create(...)` / `task_list(...)` || End session & save summary | `logInteraction(topic="...", summary="...", activity="done")` |## Memory & Context Workflow1. **Session Start**: Context is auto-injected. Skip `getContext` unless switching topics or refreshing context.2. **Logging Decisions**: Keep facts $\le 30$ words; record essential architectural decisions immediately via `logInteraction(topic, decision="...")`.3. **Session End**: Call `logInteraction(topic="...", summary="...", activity="done")` when finishing work.## Code Intelligence Rules- **Prefer `code_find` over `code_read`**: Returns line numbers only (~0 extra context tokens).- **Tight Reads**: Keep `code_read` window sizes small (`limit <= 50 lines`).- **Initial Indexing**: If `code_find` returns empty results, run `code_init(repo_path="...")` first.## Gotchas1. Avoid full-file reads unless strictly necessary — use `code_find` and `code_read` line slicing.2. Always pass the exact `topic` (repository folder name) when logging decisions.3. Keep `task_create` descriptions to single concise sentences ($\le 20$ words).

Frequently asked questions

Using Basemem