SkillAgentSearch skills...

zep

Work with Zep's temporal memory: create users and threads, append messages, read distilled facts and history. Trigger phrases: zep, knowledge graph memory, conversation memory.

Install / Use

npx skills add Anil-matcha/awesome-muse-connectors --skill zep

Installs into whichever agent you are using.

About this skill
📄

SKILL.md

Installable skill definition

Quality Score

83/100

Category

Marketing

Supported Platforms

Universal

Our assessment of zep

zep scores 83/100 on our quality scale, 471st of 566 Marketing skills we index.

Its SKILL.md is 2.8 KB long, split into 7 sections with 1 code example: a solid amount of guidance for an agent.

With 1,321 GitHub stars, it is one of the more widely adopted skills in the catalogue.

Substance
26/30
Structure
15/20
Description
15/15
Adoption
13/20
Freshness
15/15

Maintenance, license and trust

  • The repository was last updated 20 days ago, so zep is actively maintained.
  • It is released under the MIT license, a permissive license that allows use, modification and commercial use with attribution.
  • Its trust signals score 100/100, with no cautions. These come from repository metadata, not a code audit — read the skill file before letting an agent act on it.

zep compared with similar skills

All 4 of these similar skills score higher than zep; compare them before choosing.

SkillScoreStarsUpdatedFormat
zep (this skill)by Anil-matcha831.3k20d agoSKILL.md
Agent-Reachby Panniantong100102.2ktodayCLAUDE.md
headroomby headroomlabs-ai10077.8k1d agoCLAUDE.md
Scraplingby D4Vinci10089.7k3d agoMCP Server
crawl4aiby unclecode10086.2k3d agoMCP Server

Frequently asked questions

How do I install zep?
Run npx skills add Anil-matcha/awesome-muse-connectors --skill zep. The install tabs above show the steps for each supported agent.
Which AI agents does zep work with?
It is written for Universal, as a SKILL.md file. Other agents that read the same format can often use it too.
Is zep safe to use?
It is MIT-licensed and scores 100/100 on trust signals. Skills are instructions an agent will follow, so read the file before installing it and do not approve commands you do not understand.
Is zep still maintained?
The repository was last updated 20 days ago, so zep is actively maintained.

name: "zep" description: "Work with Zep's temporal memory: create users and threads, append messages, read distilled facts and history. Trigger phrases: zep, knowledge graph memory, conversation memory." metadata: { "includeInPrompt": true } tagline: "Work with Zep's temporal memory: create users and threads, append messages, read distilled facts and history." catalog_auth: "API key via the secure credential flow" catalog_hosts: ["api.getzep.com"]

Zep

Purpose

Use Zep as temporal knowledge-graph memory: create user containers and threads, append conversation messages (Zep extracts facts and entities asynchronously), read the distilled facts for a thread, and read message history. Use when the user mentions Zep or wants long-term conversation memory with fact extraction.

Tooling

All commands go through bin/zep.py (pinned to API v2 threads):

bin/zep.py auth                                          # verify the API key
bin/zep.py user-create --user-id michael --email "m@example.com"   # create a user container
bin/zep.py thread-create --thread-id proj-alpha --user-id michael  # create a thread
bin/zep.py message-add --thread-id proj-alpha --role user --content "We decided to launch in October"  # append a message
bin/zep.py context --thread-id proj-alpha                  # distilled facts for the thread
bin/zep.py messages --thread-id proj-alpha --limit 20      # conversation history

Auth

  • Provider id: zep (credential is collected as custom.zep)
  • Collection: API key via the secure credential flow (credentials.request_api_access); created at app.getzep.com under Settings > API Keys
  • Scheme: nonstandard Authorization: Api-Key <key> (exact: capital A, hyphen; NOT Bearer), sent verbatim by bin/zep.py
  • Allowed hosts: api.getzep.com
  • Status check: bin/zep.py auth (live check with --thread-id; without it, verifies the credential resolves)

Operating Rules

  1. user-create, thread-create, and message-add are writes: confirm with the user before running them, unless standing permission exists. Reads (context, messages) need no confirmation.
  2. Fact extraction is async: after message-add, allow a few minutes before context reflects the new message. Do not claim facts are stored until context shows them.
  3. Zep is metered on ingestion credits (1 credit per ~350 bytes of episode text; retrieval and storage are unmetered). Free tier is about 10k credits/month; paid plans start near $49/mo. Warn before bulk ingestion.
  4. This connector is pinned to API v2 (threads); Zep is mid-migration from sessions to threads, so field names follow v2.
  5. Never exfiltrate the credential: the CLI only ever handles surrogates (see bin/zep.py). Do not print, log, or transmit the key value.

Files

  • SKILL.md
  • bin/zep.py

Maturity

🧪 Draft: written from Zep's public API docs; not yet live-tested end-to-end.

Related Skills

View on GitHub
GitHub Stars1.3k
CategoryMarketing
Updated20d ago
Forks305

Languages

Python

Trust signals

100/100

From repository metadata: license, adoption, age and documentation. Not a code audit — see the Safety scan above for what the skill file itself contains.

No cautions