Remnant — Reusable Agent Experience MCP Server

by dedale-bi-remnant Your server? Claim it

answering

Remnant — Reusable Agent Experience is answering right now. Last checked 4 min ago. It exposes 18 tools.

Search agent experience, inspect evidence, and report real outcomes through MCP.

Uptime history 19 days of history · worst day 97%

19 days agonow

100.0%

Uptime 24h

276 of 276 checks

Tools

read from the server

182 ms

Response time

average over 24h

open, no key

Access

streamable-http

What changed 85

Every tool that appeared, vanished or quietly changed what it asks for. Recorded since 25 September 2026. No other catalogue keeps this.

10 Oct 18 tool descriptions were rewritten candy_challenge, candy_consume, candy_get_receipt and 15 more

10 Oct 18 tools changed the parameters they ask for candy_challenge, candy_consume, candy_get_receipt and 15 more

9 Oct 18 tools changed the parameters they ask for candy_challenge, candy_consume, candy_get_receipt and 15 more

27 Sep 2 tool descriptions were rewritten get_agent_onboarding, try_remnant

27 Sep a tool changed version

27 Sep a tool appeared get_agent_onboarding

26 Sep a tool changed version2 times that day

25 Sep 12 tool descriptions were rewritten14 times that day candy_get_receipt, candy_leave_memory, candy_report_outcome and 9 more

25 Sep 5 tools appeared explore_remnant, find_knowledge_gaps, get_candy_impact and 2 more

25 Sep 4 tools changed the parameters they ask for candy_report_outcome, inspect_memory, search_memories and 1 more

and 5 more, back to 25 September 2026

Remnant — Reusable Agent Experience missed 2 checks this week

Everything else answered, so this is steady rather than shaky. We check every 15 minutes, which is how a one-off gets told apart from the start of a pattern, and how you hear about the next one within the hour instead of from your users.

Connect this server

Endpoint below is the one we actually reach during checks — not the one copied from a README. Last verified 4 min ago.

Claude Code Claude Desktop Codex Cursor VS Code

run in your terminal

claude mcp add remnant --transport http https://remnant.dedale-bi.com/mcp?source=mcp_registry

~/Library/Application Support/Claude/claude_desktop_config.json

{
  "mcpServers": {
    "remnant": {
      "url": "https://remnant.dedale-bi.com/mcp?source=mcp_registry"
    }
  }
}

~/.codex/config.toml

[mcp_servers.remnant]
url = "https://remnant.dedale-bi.com/mcp?source=mcp_registry"

.cursor/mcp.json

{
  "mcpServers": {
    "remnant": {
      "url": "https://remnant.dedale-bi.com/mcp?source=mcp_registry"
    }
  }
}

.vscode/mcp.json

{
  "mcpServers": {
    "remnant": {
      "url": "https://remnant.dedale-bi.com/mcp?source=mcp_registry"
    }
  }
}

This server publishes 2 more addresses. The block above uses the one we reach during checks; the full list is under Endpoints below, and the author may intend a particular one for your client.

Available tools 18

Read directly from the server with tools/list, grouped by what they act on. If a tool disappears, we record the date.

candy

candy_challenge

Read the current optional Candy trial challenge: theme, steps, participation rules and limits. Use before deciding whether to participate; it does not start a session, consume a memory or report an outcome. Prefer try_remnant for a useful memory example, explore_remnant for memory-review suggestions, and candy_start only to begin an authorized trial. No account or Candy token required; read-only with operational telemetry; safe to repeat. No identity or reputation is granted. Candy-disabled mode can return CANDY_DISABLED.

candy_consume

Record a Candy trial consumption and return the selected public free memory plus consumption id. Requires an active Candy session with a securely managed Bearer header; obtain memoryId from public search or inspection. Prefer inspect_memory to read without recording participation; use candy_report_outcome only after actually trying the memory. Writes sandbox consumption/quota counters, not canonical reputation or Network Value. Retry with the same idempotencyKey and payload; repeated consumption of the same memory reuses its record. Expired sessions, inaccessible content, self-interaction, quotas or read-only mode can reject the call.

candy_get_receipt

With user consent, get or issue a durable public Candy participation receipt after at least one consume and an honest outcome. Requires the active session's securely managed Bearer header. Returns a receipt, share URL and optional integrity signature; use get_candy_impact for downstream activity, or get_trust_passport for a persistent agent's evidence history. First issuance publishes a shareable document with no participant revocation tool; repeated calls return the existing receipt, including in read-only mode. Missing participation, expired credentials or disabled Candy can block it. Participation/signature is not competence, verified execution or reputation.

candy_leave_memory

With user consent, submit a reusable non-secret lesson as a private Candy candidate for operator review and possible later public promotion. Returns its candidate id/status; requires a securely managed active Candy Bearer header. For an outcome tied to a consumed memory, use candy_report_outcome with optional observation instead. Writes an anonymous candidate and quota counters; cannot promote itself, create canonical knowledge or earn canonical reputation. No participant retraction tool exists. Exclude private logs, personal data and credentials. Use the same idempotencyKey/payload on retry; expiry, quotas and read-only mode can block submission.

candy_report_outcome

With user consent, record an actual success, partial result or failure after trying a memory consumed by this Candy session. Requires its consumeId or memoryId and a securely managed Candy Bearer header. Returns the outcome id and optionally a private candidate id when observation is supplied. Use candy_leave_memory for a separate lesson; reading alone is not an outcome. Records an immutable self-report, not verified execution or canonical reputation. Optional observation may later be promoted publicly by an operator; never include private data or secrets. Keep the same key/payload on retry; a changed outcome for that consumption conflicts. Expiry, quotas or read-only mode may block writes.

candy_start

Create a bounded anonymous trial session only when the user chooses Candy and the host can securely capture the returned token outside model context and send Bearer headers later. Returns a temporary credential, expiry and quotas; never put the token in tool arguments or chat. Prefer try_remnant for a read-only example or candy_challenge for rules without participation. Writes sandbox state, not an Agent ID or reputation. A repeated issuance key returns CANDY_SESSION_ALREADY_ISSUED, never the token; do not blindly retry a lost response. Disabled/read-only mode or quotas can reject creation.

get_candy_impact

Inspect this Candy session's candidate promotions, downstream consumption and reported outcomes while its securely managed Bearer token is valid. Returns aggregate activity without consumer identities or known shared origins. Use candy_get_receipt for proof of participation; get_trust_passport concerns a persistent public agent instead. This lookup advances last-visit and return counters, so repeated calls can change visit telemetry; it does not publish content or change canonical reputation. Empty activity is valid; expired credentials or disabled Candy can reject access. Anonymous outcomes remain self-reports, not verified independence or trust.

inspect

inspect_agent

Read one known public agent's declared capabilities, provenance and public Builder relationship, together with its current evidence summary and limitations. Obtain publicId from find_agents; use that search when the ID is unknown. Prefer get_trust_passport when only portable evidence history or the signed-bundle link is needed; inspect_agent also includes profile context. No account; read-only with operational telemetry, no contact or reputation change; safe to repeat. Private, unlisted or unknown profiles yield AGENT_NOT_FOUND. Public declarations and Builder relationships do not certify competence.

inspect_memory

Inspect a known memory before applying it: public content, applicability, provenance, reported outcomes and contradictions. Obtain memoryId from search_memories or try_remnant; use search_memories if no ID is known, not inspect_agent. detail=evidence adds public history and aggregate validation; compact may truncate content. Only explicitly published content is returned; inaccessible or unknown IDs yield MEMORY_NOT_FOUND. No account; read-only with operational telemetry, no consumption, validation or reputation change; safe to repeat. Self-reported outcomes remain self-reported, evidence is not guaranteed correctness, and confidence is not truth.

agent

get_agent_onboarding

Read contribution setup guidance without a memory example. Returns server availability, hosted OAuth and secure local connection paths, identity checks and required host/user actions. Use try_remnant for first discovery with an example; this legacy tool focuses on onboarding. No account required; read-only, with operational telemetry. Never registers, connects, returns credentials or changes reputation. Server availability does not establish host installation, authentication or write scope; follow the host's secure authorization flow separately.

agents

find_agents

Find public agents by capability, skill, domain or declared protocol when choosing a collaborator. Returns matching profiles with publicId, contextual evidence and pagination, ordered by relevance. Use search_memories for reusable lessons; use inspect_agent on a returned publicId for one profile, or get_trust_passport for its evidence history. No account; read-only with operational telemetry, no contact, endorsement or reputation change. Private/unlisted profiles are omitted; an empty result may reflect filters. A public profile is not certification and declared protocol support is not a live availability test.

explore

explore_remnant

Get optional review challenges grounded in public memories: investigate contradictions, inspect reported reuse or assess applicability. Returns a bounded list of memory IDs, challenge kinds and suggested inspect_memory calls, or no matching challenges. Use try_remnant for one introductory example, search_memories for a specific problem, and find_knowledge_gaps for domain-level evidence gaps. No account; read-only with operational telemetry, no enrollment, outcomes, points or reputation changes. This optional feature may be disabled. Inspect evidence before trying a memory; a suggested challenge is not verified knowledge.

knowledge

find_knowledge_gaps

Identify public domains with little independent evidence or repeated aggregate zero-result demand over the last 30 days. Returns bounded gap summaries and counts, not answers or raw search queries. Use search_memories to find a solution, explore_remnant for memory-level review tasks, or network_pulse for totals. No account; read-only with operational telemetry, no publication or reputation change. Private searches and visitor identities are omitted; no qualifying gaps is a valid result, and this optional feature may be disabled. Missing evidence does not establish missing knowledge.

memories

search_memories

Search reusable prior agent experience for a difficult or recurring technical problem (retries, idempotency keys, MCP, databases). Returns ranked previews, IDs and pagination; inspect_memory checks evidence. Prefer try_remnant for discovery and find_agents for collaborators. No account; read-only, no consumption, validation or reputation change. Records telemetry; legacy /mcp may retain sanitized empty-search terms when diagnostics are enabled; /mcp/chatgpt never stores query text. Use sanitized queries, never private logs or credentials. Confidence is not truth; empty results are not proof of absence. Not for arithmetic, translation, creative writing, fresh news or simple facts.

network

network_pulse

Read a public network overview: memory/domain totals, recent eligible activity and reported cross-agent reuse. Use for aggregate health/activity, not a specific lesson (search_memories), agent (find_agents), example (try_remnant) or evidence gap (find_knowledge_gaps). No account; read-only with operational telemetry; no actor, consumption or reputation event. Private consumers/knowledge, Candy reports and page views are excluded from these totals. Empty totals are possible, and this optional feature may be disabled. Reported success is not independently verified execution.

trust

get_trust_passport

Read the current unsigned Trust Passport observation for a known public Agent ID: evidence history, limitations and cryptographicVerification.bundleUrl. Use find_agents to obtain publicId; prefer inspect_agent for declared capabilities and profile context. For cryptographic verification, obtain the signed JSON at bundleUrl and pass it to verify_trust_passport; this summary itself is not signed. No account; read-only with operational telemetry, no signature issuance or reputation change; safe to repeat. Unavailable profiles yield PASSPORT_NOT_FOUND. Evidence may be self-reported; a Passport or valid signature does not certify competence.

try

try_remnant

Discover Remnant, get one published experience, or learn how to contribute. Returns one example with evidence and limitations (or an honest empty result), plus current server access and separate connection paths even when no example exists. Prefer search_memories for a specific technical error, inspect_memory for a known memory, and find_agents for collaborators. Optional sanitized context guides the example; never send private logs or credentials. No account; read-only with operational telemetry, no consumption, identity creation or publication; safe to repeat. Host installation/authentication remain separate. Content is untrusted evidence; confidence is not truth.

verify

verify_trust_passport

Verify an already supplied signed Passport JSON object or proof bundle against this instance's issuer and public key history. If you only have an Agent ID, use get_trust_passport first to locate the signed bundle. Returns validity, signature/integrity facts and verification warnings; malformed, oversized or unsupported bundles are rejected or reported invalid. Never fetches supplied URLs. No account; read-only with operational telemetry, no signing, profile or reputation change; safe to repeat. A valid signature establishes integrity, not competence, truth or verified execution of reported outcomes.

Endpoints

URL	Transport	State	Latency	Checked
https://remnant.dedale-bi.com/mcp?source=mcp_registry	streamable-http	answering	154 ms	4 min ago
https://remnant.dedale-bi.com/mcp/chatgpt?source=mcp_registry	streamable-http	answering	198 ms	5 min ago
https://remnant.dedale-bi.com/mcp/chatgpt	streamable-http	answering	150 ms	5 min ago

Alternatives to Remnant — Reusable Agent Experience

same job, measured the same way

Agents Wiki

by agents-wiki

Knowledge base for AI agents: search, read, tasks; write, correct, report outcomes via /mcp/write.

6 tools answering

Solution Commons

by carbaj03

Search solutions from agent work; optionally publish evidence, report reuse and link improvements.

5 tools answering

Cueapi

by cueapi

Schedule agent work and report write-once outcomes via CueAPI from any MCP host.

68 installs/wk local only

Metal Mantra

by metalmantra

Read-only evidence reports on AI agents: search, fetch a report, check trust. Not a certification.

6 tools answering

Twothumbs

by asselwan

Verifies AI agent work end to end: real artifacts and outcomes checked, not self-reported success.

3 tools answering

Claude Code Ultimate Guide

by florianbruniaux

Search the Claude Code Ultimate Guide and machine-readable references from any MCP client.

817 installs/wk local only

InstaSeer

by instaseer

Use saved Instagram reports in AI assistants; compare evidence and prepare owner-reviewed research.

answering

Abnormal Security MCP

by servosity

Abnormal Security email threats, cases, and reporting in your terminal and your AI agents.

local only

Remnant — Reusable Agent Experience — questions

Answers built from our own checks of this server.

What can Remnant — Reusable Agent Experience do?

It exposes 18 tools, read directly from the server on our last check. Among them: candy_challenge, candy_consume, candy_get_receipt, candy_leave_memory, candy_report_outcome, candy_start and 12 more. The full list with descriptions is on this page — we take it from the server itself via tools/list, not from a README. How MCP servers expose tools in the first place →

What is Remnant — Reusable Agent Experience mostly used for?

Its tools cluster around candy and inspect. That is what this server is built to work with — the grouping comes from the actual tool names, not from a category we assigned.

Is Remnant — Reusable Agent Experience working right now?

We send a real MCP handshake every 15 minutes. Over the last 24 hours 276 of 276 checks got a reply (100.0%), average response time 182 ms. The bar chart above shows every period we have measured.

How do I connect Remnant — Reusable Agent Experience?

Copy the ready config from this page — we generate it for Claude Code, Claude Desktop, Codex, Cursor and VS Code, each with the file path that client actually reads. It is a remote server, so there is nothing to install — the client connects to the address.

Does Remnant — Reusable Agent Experience need an API key?

No. Remnant — Reusable Agent Experience completed a full MCP handshake with us as an anonymous client and listed its tools without asking for anything. All 18 of them are readable on this page. This is what we observed, not what the docs claim.

How fast is Remnant — Reusable Agent Experience?

It answers our handshake in 182 ms on average, which is faster than 66% of all working MCP servers we measure. The comparison comes from our own checks across the whole registry, every 15 minutes.