Search Memory
Proactively query your Nowledge Mem knowledge base so agent replies reuse past decisions, fixes, and procedures instead of starting cold.
Overview
Search Memory is a journey-wide agent skill that queries Nowledge Mem via nmem so agents reuse past decisions and fixes whenever recall signals appear in the task.
Install
npx skills add https://github.com/nowledge-co/community --skill search-memoryWhat is this skill?
- Routes retrieval: nmem m search for durable knowledge vs nmem t search for thread/session history
- Strong signals list: prior work references, named features, reviews, regressions, implicit recall language
- Progressive thread inspection via nmem --json t show when source_thread is present
- Proactive search guidance—not only when the user says "search memory"
- JSON-first CLI contract for agent-parseable results
- Defines two retrieval paths: nmem m search (durable knowledge) and nmem t search (conversation history)
- Lists 6+ strong signals and 4 contextual signals for when to search
Adoption & trust: 613 installs on skills.sh; 91 GitHub stars; 3/3 security scanners passed (skills.sh audits).
What problem does it solve?
Your agent answers from scratch even though you already solved the bug, chose the architecture, or documented the procedure in your knowledge base.
Who is it for?
Solo builders who consistently log work in Nowledge Mem and want agents to search proactively during debugging and planning.
Skip if: Users without Nowledge Mem installed or who prefer ephemeral chat with no personal knowledge base.
When should I use this skill?
Past insights, prior fixes, named features, regressions, implicit recall language, or ambiguous debugging where Mem context would sharpen the answer.
What do I get? / Deliverables
The agent runs structured nmem searches and thread follow-ups so responses align with your stored decisions and session history.
- JSON search results incorporated into the agent reply
- Thread drill-down when source_thread is returned
Recommended Skills
Journey fit
Useful at every journey phase - explore requirements and options before committing to a direction.
Where it fits
User reports a regression in a named connector; agent searches Mem for the prior fix and rationale.
Architecture discussion triggers a search for stored API conventions the user documented last sprint.
Pre-release review searches for open decisions and checklist items saved after the last launch.
Support-style question matches a solved incident thread via nmem t search.
Scoping a feature references "the pattern we used"—agent pulls the earlier scope notes from durable memory.
How it compares
Procedural recall routing for nmem—not a generic web search or vector-db setup skill.
Common Questions / FAQ
Who is search-memory for?
Developers and founders using Nowledge Mem who want coding agents to pull past insights, fixes, and thread context automatically.
When should I use search-memory?
During operate/iterate when resuming bugs or incidents; during build when architecture matches past decisions; during ship review or grow/content when aligning docs; whenever the user references "like before" or a named subsystem.
Is search-memory safe to install?
It instructs local nmem CLI calls that read your knowledge base; review the Security Audits panel on this Prism page and treat memory contents as sensitive data.
SKILL.md
READMESKILL.md - Search Memory
# Search Memory > AI-powered search across your personal knowledge base using Nowledge Mem. ## When to Use **Strong signals — search when:** - the user references previous work, a prior fix, or an earlier decision - the task resumes a named feature, bug, refactor, incident, or subsystem - the task is a review, regression, release, docs-alignment, or connector-behavior question - a debugging pattern resembles something solved earlier - the user asks for rationale, preferences, procedures, or recurring workflow details - the user uses implicit recall language: "that approach", "like before", "the pattern we used" **Contextual signals — consider searching when:** - complex debugging where prior context would narrow the search space - architecture discussion that may intersect with past decisions - domain-specific conventions the user has established before - the current result is ambiguous and past context would make the answer sharper ## Retrieval Routing 1. Start with `nmem --json m search` for durable knowledge. 2. Use `nmem --json t search` when the user is really asking about a prior conversation or exact session history. 3. If a result includes `source_thread`, inspect it progressively with `nmem --json t show <thread_id> --limit 8 --offset 0 --content-limit 1200`. 4. Prefer the smallest retrieval surface that answers the question. For continuation-heavy engineering work, search near the start of the task. Do not wait for the user to literally ask for memory search. If the host already knows the active project or agent lane, add `--space "<space name>"` to these commands. ## Native Connector These skills work in any agent via CLI. For auto-recall, auto-capture, and graph tools, check if your agent has a native Nowledge Mem connector — run the `check-integration` skill or see https://mem.nowledge.co/docs/integrations