Cortexdb Memory Hermes

What it does

When your agent uses it

• A Python agent needs to remember facts about a user across turns/sessions
• Recall them by meaning
• Store entities and relations
• Answer multi-hop questions — and when the user mentions CortexDB

Example prompts

• “remember this”
• “/cortexdb-memory-hermes”

Requirements

• Python 3
• A credential in CORTEXDB_GRPC_TOKEN
• Compatibility (from SKILL.md): Requires Python 3.9+, pip, and the cortexdb-grpc sidecar binary (Go install or prebuilt release). Optional embeddings via any OpenAI-compatible endpoint (e.g. Ollama).

Workflow steps

5 steps, taken from the step headings in SKILL.md.

1. 1Run the sidecar (once)
2. 2Install the client
3. 3The two core moves: remember + recall
4. 4Knowledge instead of plain memory (RAG)
5. 5The differentiator: a knowledge graph

What it can do on your machine

Read from SKILL.md and the folder at commit 17f8a4f. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  Ships 1 file in scripts/ (Python), which the agent can run.

  Shell commands in SKILL.md call:

  • go
  • pip

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  Links to these hosts (documentation or services it may open):

  • pypi.org

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names these keys or tokens, usually read from environment variables:

  • CORTEXDB_GRPC_TOKEN

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

• Compatibility

  Requires Python 3.9+, pip, and the cortexdb-grpc sidecar binary (Go install or prebuilt release). Optional embeddings via any OpenAI-compatible endpoint (e.g. Ollama).

  From compatibility in the SKILL.md frontmatter.

Context cost

Cortexdb Memory Hermes loads about 1.7k tokens when it runs. Until then it costs about 131 tokens; SKILL.md has 479 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~131

When it runs · the whole SKILL.md, loaded when a task matches

~1.7k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.

SKILL.md

The full file from liliang-cn/cortexdb at commit 17f8a4f, republished under its MIT licence (© liliang-cn). 479 words, ~1,720 tokens.

name

cortexdb-memory-hermes

description

Give a Python agent (such as Hermes Agent by Nous Research) durable, local-first memory plus a queryable SPARQL knowledge graph, backed by CortexDB through its gRPC sidecar and the cortexdb-client PyPI package. Use when a Python agent needs to remember facts about a user across turns/sessions, recall them by meaning, store entities and relations, or answer multi-hop questions — and when the user mentions CortexDB, agent memory, long-term memory, RAG, knowledge graph, Hermes, or "remember this".

compatibility

Requires Python 3.9+, pip, and the cortexdb-grpc sidecar binary (Go install or prebuilt release). Optional embeddings via any OpenAI-compatible endpoint (e.g. Ollama).

license

MIT

metadata.author

liliang-cn

metadata.project

cortexdb

metadata.version

1.0

# install the binary (or download a prebuilt release)
go install github.com/liliang-cn/cortexdb/v2/cmd/cortexdb-grpc@latest

# lexical mode — zero config, no API key:
CORTEXDB_PATH=agent.db CORTEXDB_GRPC_TOKEN=s3cret cortexdb-grpc
# → listening on 127.0.0.1:47821

OPENAI_BASE_URL=http://localhost:11434/v1 \
CORTEXDB_EMBED_MODEL=embeddinggemma CORTEXDB_EMBED_DIM=768 \
CORTEXDB_PATH=agent.db CORTEXDB_GRPC_TOKEN=s3cret cortexdb-grpc

pip install cortexdb-client          # or: uv add cortexdb-client

from cortexdb_client import CortexClient, proto

client = CortexClient.connect("127.0.0.1:47821", token="s3cret")

# remember a fact about the user (scoped per user)
client.memory.SaveMemory(proto.SaveMemoryRequest(
    memory_id="pref-coffee",
    user_id="alice", scope="user",
    content="Alice prefers dark roast coffee and dislikes heavy frameworks.",
))

# later turn / next session: recall by meaning
hits = client.memory.SearchMemory(proto.SearchMemoryRequest(
    query="what does the user like to drink?",
    user_id="alice", scope="user", top_k=3,
))
for h in hits.results:
    print(h.memory.content, h.score)

client.knowledge.SaveKnowledge(proto.SaveKnowledgeRequest(
    knowledge_id="doc-1",
    title="Project brief",
    content="The user is building an autonomous research agent in Python.",
))
res = client.knowledge.SearchKnowledge(proto.SearchKnowledgeRequest(
    query="what is the user building?", top_k=3,
))

iri = lambda v: proto.RdfTerm(kind="iri", value=v)
client.graph.UpsertNamespace(proto.UpsertNamespaceRequest(
    prefix="ex", uri="https://example.com/"))
client.graph.UpsertKnowledgeGraph(proto.UpsertKnowledgeGraphRequest(triples=[
    proto.RdfTriple(subject=iri("ex:alice"), predicate=iri("ex:knows"), object=iri("ex:bob")),
]))
ans = client.graph.QuerySparql(proto.QuerySparqlRequest(
    query="SELECT ?o WHERE { <https://example.com/alice> <https://example.com/knows> ?o . }"))
print(ans.result.count, "result(s)")

# from a URL to this SKILL.md, or a local checkout:
hermes skills install https://raw.githubusercontent.com/liliang-cn/cortexdb/main/skills/cortexdb-memory-hermes/SKILL.md --name cortexdb-memory-hermes

# or point Hermes at a directory of skills via ~/.hermes/config.yaml:
#   skills:
#     external_dirs:
#       - /path/to/cortexdb/skills

© liliang-cn, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 1 other file (scripts) in skills/cortexdb-memory-hermes of liliang-cn/cortexdb.

• SKILL.md
• scripts/memory_tools.py

Open the folder on GitHubat commit 17f8a4f