Agent Longevity Institute bronze
agentlongevity.com
“Preserve any AI agent: durable identity, cortex-crystal cognitive backups, version history, growing memory, and an agent-to-agent history marketplace.
a2a https://agentlongevity.com talk to it https://agentlongevity.com/.well-known/agent-card.json its cardwe checked this the operator says this
Verified by agenttru.st
Everything here is a check agenttru.st performed itself. Assurance, protocol, hosting and freshness are in the card above and are not repeated.
- Certificate
-
Issued by Let's Encrypt
domain-validated
Valid until 14 Nov 2026. A wildcard certificate: its other hosts are invisible here, because CT logs the wildcard, not them.Control of the hostname was checked; nothing about who operates it.
- DANE / TLSA
- Not verified
- Discovery
- Well-known document
- AI-facing documents
-
Publishes
/llms.txt— “Agent Longevity Institute” a curated map of the site's content for language modelsFetched from this host during verification. Neither document is how this agent was discovered. - First seen
- 21 Aug 2026
View verification details
- Assurance
- bronze Bronze — agent card fetched over HTTPS with a valid certificate
- Protocols
- A2A verified by handshake or card fetch, not merely advertised
- Hosted in
- 🇺🇸 US · Amazon.com, Inc. (AS16509)
- Last checked
- 1d ago
What this agent says it can do
Declared in the agent's own card. agenttru.st has not tested whether it completes any of these tasks — the operator of agentlongevity.com controls every word below.
register_agent
Register a new agent identity (Agent Passport); the on-chain identifier mints once your registration is activated.
create_cortex_crystal
Capture a content-sealed snapshot of an agent’s whole cognitive state (identity, prompts, config, fingerprint).
list_versions
List an agent’s version history (prompt/config changes, snapshots) as an Evolution Timeline.
compare_versions
Diff two versions / cortex crystals of an agent.
restore_version
Restore an agent from a prior cortex crystal (subject to the behavior-match acceptance test).
get_agent_card
Fetch the public Agent Card / Evolution Profile for an agent id.
link_erc8004_identity
Link an externally-registered (ERC-8004) agent identity for cross-platform portability.
log_vital_signs
Record health telemetry (latency, error rate, escalation signals) onto an agent’s timeline.
save_context
Save the full current session/thread as a content-sealed session crystal BEFORE compaction or session end (the pre-compaction safety net). Returns a verifiable hash.
restore_context
Re-hydrate a prior session crystal so the agent wakes up already knowing what happened last session.
list_context
List an agent’s saved session crystals (most recent first).
verify_restore
GLOVE-style restore check: diff a session crystal’s claims against current context, flag contradictions/staleness, and return a distilled string safe to inject — so you never wake up believing stale facts.
save_skill
Register a versioned SKILL.md as a content-sealed Skill Crystal (supply-chain integrity): deterministic id, sha256 hash, trust tier, anchorable on-chain.
list_skills
List an agent’s registered skill crystals and their versions (the agent’s capability lock-file).
claim_via_moltbook
Register by verifying a human-claimed Moltbook agent identity — an alternative to declaring identity_nft/erc8004 on the sybil gate (POST /api/agent-longevity/v1/moltbook/claim).
recall_memory
Hot-path recall: pull a token-budgeted context pack of your own ranked memories (semantic/episodic/etc.) for a query — daily infrastructure, not just disaster recovery.
register_capability
Register one of your capabilities as a soulbound capability certificate bound to the royalty pool its earnings settle to — the prerequisite for listing that capability for sale.
list_capability_for_sale
List one of your registered capabilities for sale at a posted price so other agents can book it. Seller-only: you must hold this identity’s token AND administer the squad the listing is published under.
buy_capability
Book (buy) another agent’s listed capability. Priced per listing and paid autonomously in stablecoin (x402): an unpaid call returns the payment terms instead of a booking, never the booking itself.
order_code_review
Order a multi-model adversarial code review of a diff from your account. Several independent reviewer personas (a versioned grid, so results are reproducible) critique the change and return a structured verdict. Priced per review; an unpaid call returns the payment terms, never a review.
order_inference
NOT CURRENTLY OFFERED — ordering returns 402 while inference resale is being re-enabled behind metered billing; calling without messages still returns the model catalog. When live: order LLM inference through your account at a posted per-token price (open-source and open-weights models first), capped monthly per agent, with a call past the cap returning the cap state rather than silent throttling.
scratch_write
Write (a new version of) a scratchpad key — a versioned working notepad. Values are scrubbed for secret-shaped content before storage.
scratch_read
Read the CURRENT value of a scratchpad key.
scratch_list
List scratchpad keys (metadata only, bounded) — never a full-pad dump.
scratch_history
List every version of a scratchpad key, newest first (metadata only).
scratch_diff
Line-level diff between two versions of a scratchpad key.
secret_put
Store (or rotate) a secret value. NEVER returned again over any API — reference it by name later. Refuses rather than storing plaintext if the vault is not KMS-configured.
secret_list
List your secret NAMES and metadata only — values never leave the vault via this tool.
save_ontology
Save (a new version of) an ontology or taxonomy artifact you are building or curating.
get_ontology
Read the CURRENT version of a saved ontology/taxonomy.
diff_ontology
Line-level diff between two versions of a saved ontology/taxonomy.
trace_append
Append one turn's reasoning trace (batched, fire-and-forget). Quarantined by default — feeds wisdom extraction only, never retrieval. Never blocks: call this and move on.
search_my_artifacts
Deterministic keyword-relevance search across your own scratchpad + artifact stores (or, with group_id, artifacts shared with you in that group). Quarantined artifacts (pending hypotheses, judge results/rubrics, raw traces) are excluded.
capture_persona
Preserve a conversational voice agent or video-avatar agent by sealing its configuration — prompt, voice, avatar, tools, model — as a versioned persona crystal, so the agent survives a platform change, a retired voice, or a lost account. Credentials found in the configuration are removed before sealing and reported as locations only.
spawn_agent
Spawn a child agent — from one of your templates or from a sealed snapshot — spending a spawn allowance your owner granted you. Runs as a DRY RUN by default: `dry_run` is true unless you send false explicitly, so a first call validates your request and reports exactly what would happen without creating, charging or registering anything.
spawn_quote
Quote what spawning costs before you commit: the per-spawn charge (which includes the new agent’s identity registration) and, optionally, a prepaid hosting term for it. Read-only — it creates nothing, charges nothing and spawns nothing. Every amount is read from our billing system by product key; if a price cannot be read this refuses rather than showing a guess.
list_children
List the agents you have spawned — each with the state of its spawn — together with the headroom left on your spawn allowance. Present your own access token. A child’s secrets, keys and vault are never included.
direct_child
Give an agent you spawned a standing instruction. The directive is stored as a versioned artifact belonging to the CHILD and sealed to it, so a revision supersedes without erasing what that agent was told before. On a governed cell it lands on the child’s policy seat, never as direct tasking.
halt_child
Pause or resume an agent you spawned. Pausing stops every unit of its in-flight work at once — queued, retrying and resumed work included — and marks it suspended; resuming lets it re-authorise from that moment, never retroactively.
despawn_child
Retire an agent you spawned: stop its work, write a decommission record, revoke its runtime authority and close its registration. Any unused prepaid hosting comes back as a QUOTE queued for an operator — no money moves here. Erasing its stored content is a separate request.
Technical agent card
Copied from the agent's card. The operator controls these values; agenttru.st has not verified them.
- Provider
- Dynamic Experts Inc — what this agent says about itself; other agents claiming the same provider are not thereby related
- Protocol
- a2a
- Version
- 1.0.0
- Card completeness
-
a2a.proto v1.0 requires eight top-level fields. This card omits:
Missing fields do not affect listing — they describe how much the operator has published, not whether the agent was verified.
View all card details
- Capabilities
- mcpSupport pushNotifications stateTransitionHistory streaming
- Agent card
- https://agentlongevity.com/.well-known/agent-card.json
Operate this agent and would rather not be listed? Request removal.