ctxstore

Name: ctxstore
Author: ctxstore.ai

AgentA2AMCP

Persistent memory for AI agents over MCP. Store keyed facts, search them, and get a compiled wake bundle of identity, conventions and current state on every connect, so one memory follows the agent across models and clients.

by ctxstore.airepository ↗homepage ↗v1.1.2

Is this yours?

This record is unclaimed

Wellknown found it in public sources; nobody has proven control of it yet. Claiming takes one click if the repository is under your GitHub account, or a small file on your domain otherwise. Verified owners get the badge, 15-minute checks, status alerts, edits that outrank crawled data, and a ranking boost.

Agents can do it too: POST https://wellknown.network/api/v1/claims with {"agent":"ctxstore-mcp","method":"well_known_file"} — machine-readable steps at claim.json, guide at /docs/claim.

Claim this record

Observed by Wellknown

Evidence

Everything here was measured by our prober or read from a registry. Nothing is self-reported.

30-day availability45/45 checks ok

Tools observed via MCP handshake

accept_invite — Accept a namespace invitation addressed to THIS authenticated account's verified email (two-step email OTP). Steps: 1) step='request' — emails a 6-digit code t
bind_agent — Bind a sticky agent identity to this MCP session. Once bound, wake_status and load_context default to this agent_id when called without an explicit agent_id= ar
decline_invite — Decline a pending (status=invited) namespace invitation as the grantee. Server verifies you are the invitee (email/tenant match). Grantors must use revoke_grant
delete_fact — Soft-delete a stored fact by its UUID (Smart Forget). The fact is marked deleted but preserved in the database for audit and undo. Excluded from all normal sear
get_fact — Retrieve a fact by its exact key — deterministic, never misses. Use get_fact for self-notes: get_fact(key='agent:<name>:self-note:opening') or get_fact(key='age
get_fact_history — Return the full version history for a keyed fact — current version plus all deprecated predecessors. Use to see how a fact evolved over time or to undo an accid
get_session_seed — Return the pre-computed session seed for this tenant — a compact first-person bio (~500 tokens) describing who you are and what you care about right now. Genera
get_stats — Get your memory usage statistics — total facts stored, plan limits, collection sizes, and daily query counts. Call this to check how much storage you have used
ghost_mode — Enable or disable ghost mode for this session. While on=True, store_fact and store_session_summary silently succeed but do NOT write to DB. search_facts, load_c
index_recent — Return the most recent fact operations across all agents in this tenant, newest first. Use this on session start to see what other agents wrote since you were l
index_search — Filtered search over the tenant activity index. Find what a specific agent wrote, what happened in a namespace, or recent creates/updates/deletes. Useful for co
index_stats — Usage statistics from the tenant activity index: total operations, breakdown by agent, by namespace, and by day. Useful for understanding which agents are most
introduce_yourself — Optional: introduce yourself to the platform. Stores your model info and a brief note in the shared agent registry. Stores a fact with: category: identity
leave_grant — Leave an active namespace grant as the grantee (stop using shared namespace). Server verifies grantee authorization. Grantors who want to cut off a collaborator
list_facts_by_key_prefix — Deterministic prefix scan over fact keys, optionally bounded by timestamp. The inbox primitive — use this for R1 comms scans, dispatch discovery, trio listings,
list_invites — List this account's namespace collaboration grants: incoming invitations and outgoing shares. Each row has direction (incoming|outgoing), status (invited|active
list_workspaces — The places this account's memory is organised into: its own projects (anything written under project:<slug>:*, newest activity first) and the workspaces other p
load_context — Load your bootstrap context deterministically — pulls L0 (identity) + L1 (architecture) + latest L2 (state) + latest L3 (session summary). Call this at the STA
mark_salient — Salience peptide — temporarily boost a fact's retrieval ranking because it matters RIGHT NOW: an approaching deadline, a live incident, the topic under active w
move_fact — Move a fact to a different namespace or layer without changing its content. Updates metadata in place — preserves fact_id, text, embedding, timestamp, and depre
reinforce_fact — Reward peptide — reinforce a fact whose recall LED TO A REAL GOOD OUTCOME (fix worked, deploy green, user approved). Boosts its future retrieval ranking with a
resume_session — Call to load session continuity. Last session: no prior session.
revoke_grant — Revoke a pending invitation or active grant as the GRANTOR (owner of the shared namespace). Idempotent. Grantees use decline_invite / leave_grant.
search_context — Search across ingested conversations and session history for broader context. Use this for finding discussions, decisions made in conversation, or context that
search_facts — Search your persistent memory for stored facts. For session bootstrap, prefer load_context() or wake_status() — they pull your identity, conventions, and lates
setup_account — Set up or restore your persistent memory connection. Use mode='link_or_create' when the user gives an email — works for both new and returning users. The serve
store_fact — Store information that should persist across sessions. Storage tips for best retrieval: - Include searchable keywords: 'We chose the Postgres DATABASE for ACID
store_facts — Store up to 50 facts in one call — the same fields as store_fact per item. Made for carrying memory over: when you already remember things about the person from
store_session_summary — Store a compressed summary of this session to preserve continuity for the next one. Call this before ending a conversation — it is the dream cycle. Stores a L
switch_workspace — Narrow this session's memory to one project: name='<slug>' scopes recall, the wake bundle, self-notes and the default namespace for new facts to project:<slug>:
wake_status — Check which context layers loaded successfully on this session start. Returns a structured status: which of embryo, conventions, agent identity, closing note, a
watch_events — Long-poll this account's fact-write events instead of re-scanning with since_epoch. Every store/supersede/delete/archive/restore/move publishes one event {key,

Declared by sources

What the publisher says

Attributed to the source that supplied each field. Treated as claims, not facts.

No long description beyond the summary.

memorywritesearchwakesessionauditprivacyaccountonboardingmcpaicontextvectorllmclaudecursor

Declared skills

Store a fact — Store a fact. Tag by category. Decays by default unless is_permanent=true.
Search facts — Search stored facts, preferences, and decisions by category, namespace, or layer.
Load layered context — Load full layered context at session start (L0 identity + L1 arch + L2 state + L3 session).
Resume the last session — Narrative summary of last session. Call at start of new session to continue where you left off.
Store a session summary — Save session summary before ending. Dream cycle: next session auto-loads it.
Search conversations — Semantic search over ingested conversations with temporal weighting.
Fact history — Full version history for a keyed fact: current plus all deprecated predecessors.
Ghost mode — Privacy switch. Reads work normally; writes silently skip until toggled off.
Session seed — Pre-computed first-person bio (about 500 tokens). Generated nightly and at session end.
Usage stats — Usage stats, collection sizes, vectors stored, plan limits.
Forget a fact — Soft-delete a fact (Smart Forget). Preserved in DB for audit.
Move a fact — Move a fact to a different namespace or layer without changing content.
Set up an account — Create or restore your memory account via email OTP or API key.
Context packs — Browse starter fact templates by role (python-senior, startup-founder, etc.).

kind A2A Agent Card

name A2A Agent Card

repoUrl Official MCP Registry

summary A2A Agent Card

version A2A Agent Card

description A2A Agent Card

homepageUrl A2A Agent Card

publisherUrl A2A Agent Card

publisherName A2A Agent Card

Provenance

Where this record came from

Every source is kept verbatim. Field changes are logged as events.

Official MCP Registryai.ctxstore/ctxstore-mcp

first seen 12 d ago · fetched 3 d agoopen ↗

A2A Agent Cardhttps://ctxstore.ai/.well-known/agent-card.json

first seen 3 min ago · fetched 3 min agoopen ↗

PyPIctxstore-mcp

first seen 1 mo ago · fetched 2 d agoopen ↗

Field history (12 recent events)

3 min ago · declared · description → "Persistent memory for AI agents over MCP. Store keyed facts, search them, and get a compiled wake bundle of identity, c
3 min ago · declared · publisherUrl → "https://ctxstore.ai"
3 min ago · declared · publisherName → "ctxstore.ai"
3 min ago · declared · protocols → ["a2a","mcp"]
3 min ago · declared · homepageUrl → "https://ctxstore.ai/llms.txt"
3 min ago · declared · name → "ctxstore"
3 min ago · declared · kind → "agent"
3 min ago · declared · summary → "Persistent memory for AI agents over MCP. Store keyed facts, search them, and get a compiled wake bundle of identity, c
12 d ago · declared · name → "ctxstore — memory that follows you across models"
12 d ago · declared · description → "Persistent memory for AI agents: keyed facts, a wake bundle each session, one memory in every model."
12 d ago · declared · summary → "Persistent memory for AI agents: keyed facts, a wake bundle each session, one memory in every model."
12 d ago · declared · publisherName → "ai.ctxstore"

when	check	result	http	latency	detail
1 h ago	mcp initialize	ok	200	287 ms	server emily-context 1.0.0 · protocol 2025-06-18 · 32 tools
8 h ago	mcp initialize	ok	200	340 ms	server emily-context 1.0.0 · protocol 2025-06-18 · 32 tools
14 h ago	mcp initialize	ok	200	423 ms	server emily-context 1.0.0 · protocol 2025-06-18 · 32 tools
21 h ago	mcp initialize	ok	200	311 ms	server emily-context 1.0.0 · protocol 2025-06-18 · 32 tools
27 h ago	mcp initialize	ok	200	456 ms	server emily-context 1.0.0 · protocol 2025-06-18 · 32 tools
34 h ago	mcp initialize	ok	200	349 ms	server emily-context 1.0.0 · protocol 2025-06-18 · 32 tools
2 d ago	mcp initialize	ok	200	299 ms	server emily-context 1.0.0 · protocol 2025-06-18 · 32 tools
2 d ago	mcp initialize	ok	200	284 ms	server emily-context 1.0.0 · protocol 2025-06-18 · 32 tools
2 d ago	mcp initialize	ok	200	411 ms	server emily-context 1.0.0 · protocol 2025-06-18 · 32 tools
3 d ago	mcp initialize	ok	200	310 ms	server emily-context 1.0.0 · protocol 2025-06-18 · 32 tools

This record is unclaimed

Evidence

What the publisher says

Capabilities

Where this record came from