Skip to content
Back to skills

Nowledge Mem

ASecurity

Cross-tool knowledge for Amp. Read startup Context Bundle, search past decisions and conversations, save durable insights, and capture sessions as searchable threads with Nowledge Mem.

  • 182 stars
  • 0 votes
  • 0 copies
  • 1 view
  • Added September 24, 2026
ai-agentsgodebuggingdocumentation

Security analysis

A100/100

Scanned September 24, 2026

npx -y skills add nowledge-co/community --skill nowledge-mem --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Nowledge Mem?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Nowledge Mem
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/nowledge-co-nowledge-mem-community/badge)](https://www.skillsdirectory.com/skills/nowledge-co-nowledge-mem-community)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: nowledge-mem
description: Cross-tool knowledge for Amp. Read startup Context Bundle, search past decisions and conversations, save durable insights, and capture sessions as searchable threads with Nowledge Mem.
---

# Nowledge Mem

You have Nowledge Mem tools (`nowledge_mem_*`) for cross-tool knowledge management. Use them proactively.

## At session start

Call `nowledge_mem_context_bundle` when identity, scope, or rules may matter. It returns owner identity, resolved AI Identity, active scope, active rules, Working Memory, and KFS paths in one stable contract. Use `nowledge_mem_working_memory` only for a lightweight daily briefing or as a fallback when the Context Bundle is unavailable.

## When to search (`nowledge_mem_search`)

Search proactively when past context would improve the answer. Strong signals:

- The user references previous work, a prior fix, or an earlier decision
- The task resumes a named feature, bug, refactor, or subsystem
- A debugging pattern resembles something solved earlier
- The user asks for rationale, preferences, procedures, or recurring workflow details
- The user uses recall language: "that approach", "like before", "the pattern we used"

Do not search when the topic is fundamentally new, when the answer is generic documentation, or when the user explicitly wants a fresh perspective.

## When to search threads (`nowledge_mem_thread_search`)

- The user asks about a prior conversation or exact session history
- A memory result references a source thread

## When to save or update

Save proactively when the conversation produces a durable fact, preference, decision, plan, procedure, learning, event, or important context. Do not wait to be asked. Search first to check for related memories:

- If a related memory exists, call `nowledge_mem_update` to refine it
- If the insight is genuinely new, call `nowledge_mem_save`

Good candidates: decisions with rationale, repeatable procedures, lessons from debugging, durable preferences, plans future sessions will resume. Skip routine fixes, in-progress work, and generic Q&A answerable from documentation.

## When to save the session (`nowledge_mem_save_thread`)

- The user asks to save the conversation or "remember this session"
- A long, productive session is wrapping up
- The conversation produced decisions or context worth preserving as a full thread

Session capture also happens automatically at the end of each agent turn; the tool is the explicit, on-demand path and is idempotent.

## When to save a handoff (`nowledge_mem_save_handoff`)

Use a handoff when you want a curated summary rather than the full transcript: provide a topic and a structured summary (Goal, Decisions, Key files, Risks, Next steps).

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.