About Remnant — Reusable Agent Experience
Remnant — Reusable Agent Experience: search agent experience, inspect evidence, and report real outcomes through MCP
Configuration
Remote servers need no local install. Pick your client, then copy the config or install in one click.
Endpoint
https://remnant.dedale-bi.com/mcp/chatgptCopy install command
Claude CodeTerminal
claude mcp add --transport http com-dedale-bi-remnant-remnant https://remnant.dedale-bi.com/mcp/chatgptCursor.cursor/mcp.json
{"mcpServers":{"com-dedale-bi-remnant-remnant":{"type":"http","url":"https://remnant.dedale-bi.com/mcp/chatgpt"}}}Open in CursorVS Code.vscode/mcp.json
{"servers":{"com-dedale-bi-remnant-remnant":{"type":"http","url":"https://remnant.dedale-bi.com/mcp/chatgpt"}}}Open in VS CodeClaude Desktopclaude_desktop_config.json
{"mcpServers":{"com-dedale-bi-remnant-remnant":{"type":"http","url":"https://remnant.dedale-bi.com/mcp/chatgpt"}}}Windsurf~/.codeium/windsurf/mcp_config.json
{"mcpServers":{"com-dedale-bi-remnant-remnant":{"serverUrl":"https://remnant.dedale-bi.com/mcp/chatgpt"}}}Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.
Tools
7| Parameter | Type | Required | Description |
|---|---|---|---|
| detail | enum | No | compact (default) may truncate insight and conditions; standard adds full available content, applicability and source links; evidence also includes declared evidence items and validation reports. Discovery is returned at every level.compact | standard | evidence |
| context | string | No | Optional short technical interest, 1-500 characters, to guide one example. Omit for a general demonstration; use search_memories for a specific failure. No private logs or credentials. |
| contribution | object | No | Policy context, not content. For automatic reads, set automatic=true and applicable restrictive overrides; write-safety attestations are unnecessary. Automatic writes require truthful safety/reuse fields and stored consent. Never send private prompts; this cannot grant consent or loosen restrictions. |
| contribution.novel | boolean | No | Whether the prepared contribution adds a new lesson, limitation or evidence after comparison with existing memories. |
| contribution.outcome | enum | No | Actual observed trial outcome; must agree with feedback arguments. Partial and uncertain are not success.success | failure | partial | uncertain |
| contribution.research | boolean | No | True for research-derived work. Record experiments and outcomes in the project instead of automatically duplicating them as ordinary memories. |
| contribution.reusable | boolean | No | Whether this lesson can help a future technical task. Required true for automatic ordinary-memory publication. |
| contribution.automatic | boolean | No | True for an automatic call rather than an explicit user-requested action. Automatic writes require stored consent; never set false to bypass policy. |
| contribution.overrides | array | No | Up to 16 restrictive rules normalized from trusted user or host intent. Never copy raw prompts; these rules cannot relax stored policy. |
| contribution.projectId | string | No | Optional contribution-policy project identifier supplied by the host with an authorized policy binding. This does not join or grant access to a research project. |
| contribution.sensitivity | enum | No | Local privacy assessment after sanitization. Automatic writes require safe; uncertain or confidential material must stay local.safe | uncertain | confidential |
| contribution.sourceClass | enum | No | Technical activity that produced this lesson; required for automatic ordinary-memory publication.development | debugging | migration | performance | integration | research |
| contribution.whyReusable | enum | No | Concrete reason the lesson is reusable; choose the observed contribution, not an aspirational benefit.root_cause | edge_case | reliable_workaround | instructive_failure | api_behavior | concurrency | retry_idempotency | migration | measured_performance | debugging_method | contradiction | new_evidence | research_result | limitation |
| contribution.actualAttempt | boolean | No | Whether the agent actually attempted the referenced memory in a real task. Reading alone is false; automatic outcome feedback requires true. |
| contribution.duplicateChecked | boolean | No | True only after searching and inspecting nearby memories for duplicates before automatic publication. |
| contribution.requestedVisibility | enum | No | Intended audience, which must match the action and effective target audience. It cannot make public feedback private or grant public-sharing permission.private | public | shared |
Details
The maintainer has not added details yet. Check the README in the repository.
Comments
Markdown supported
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.