onedevatlasBeta
Browse the atlas
AllCAPUI5ABAPBTP
OrganisationIndependent

ARC MCP

8maintains
GitHub@arc-mcplast push

Capabilities4

Provides 4
Agent access to your codeABAPBTPAgent access to your codecapabilityLetting a coding agent read the project or the system you actually have, reach its model, its objects and its checks, and act on it. In file-based environments that works out of the box; complex systems need dedicated access capabilities. Whatever the environment, the agent also needs project- and framework-specific context, so it understands the thing in front of it rather than guessing from what it remembers of its training data. The transport is incidental: an MCP server, a language server or a CLI answer the same question, and what differs between them is how much of the real thing the agent can see. Distinct from [MCP servers](capability:mcp-servers), which is a protocol adapter on your own running service, and from [agent guidance](capability:agent-guidance), which is knowledge handed over as files rather than read from the thing itself.Agent guidanceABAPBTPAgent guidancecapabilityPackaged knowledge you hand a coding agent so it gets a framework right unprompted — skills, rule files, instruction packs. Distinct from agent tooling, which is about building agentic features into your application, and from [agent access to your code](capability:agent-code-access), where the agent reads the real project instead of being handed a file about it.Editor supportABAPBTPEditor supportcapabilityWorking in source code with the editor helping rather than treating it as text — completion, go-to-definition, inline errors, refactoring and preview.Tool governanceABAPBTPTool governancecapabilityControlling what an agent may call, with which authorisations, and leaving an audit trail of what it did. The authorisation question that MCP exposure raises but does not answer.
Demonstrates 1
Agent access to your codeABAPBTPAgent access to your codecapabilityLetting a coding agent read the project or the system you actually have, reach its model, its objects and its checks, and act on it. In file-based environments that works out of the box; complex systems need dedicated access capabilities. Whatever the environment, the agent also needs project- and framework-specific context, so it understands the thing in front of it rather than guessing from what it remembers of its training data. The transport is incidental: an MCP server, a language server or a CLI answer the same question, and what differs between them is how much of the real thing the agent can see. Distinct from [MCP servers](capability:mcp-servers), which is a protocol adapter on your own running service, and from [agent guidance](capability:agent-guidance), which is knowledge handed over as files rather than read from the thing itself.

Entries8

Group by8 of 8 entries
EntrySignals
Maintains7
@arc-mcp/adt-lsv0.7.0
Functional boundary

TypeScript SDK over SAP's headless adt-ls language server

arc-1v1.9.0
none

MCP server over the ADT REST API for ABAP objects

ARC-1 Skills
Functional boundary

Twenty-two agent skills shipped with arc-1

arc-1-lspv0.0.1
Functional boundary

Edition of ARC-1 that runs ADT calls through adt-ls

arc-mcp-hub
Archived+1

Hub on SAP BTP putting several arc-1 instances behind one login

arc1-abap-bridge
Functional boundary

VS Code bridge between arc-1 and ADT for VS Code

arc1-adt-abap-mcp-ext
Functional boundary

Eclipse add-on enabling the ADT MCP server with read-only tools

Maintains: samples, sketches and references1
arc-1-extension-sample
No license+1

Sample arc-1 extension in TypeScript with ten custom tools

Writing and topics1

IN TOPICS

The atlas is now in betaTopic, edition 1names arc-1

onedevatlas is open for browsing. Find tools for SAP development, compare their documented capabilities, ask your AI assistant, and join a flight during TechEd week.

Details

History13
  • ARC-1 Skillsprovides.agent-guidance.facets → installs_into: six clients the manual-install table names; system_reach: You connect itcapability:agent-guidance widened installs_into to the ABAP door and declared system_reach there. The skills README read on 2026-10-15 names each client one at a time in its manual-install table, with a directory per client, which is why each value carries its own row; the plugin routes it documents beside that are for Claude Code and Agent Plugins, and the skills command line it offers states "any agent" rather than a list. system_reach rests on the prerequisite the README already states, which the entry's own warning repeats: the server is required for every skill that touches ABAP, and the reader connects it.
  • arc-1description.when_not_to_use → removedThe licensing side is its explicit SAP API Policy relation, sourced to its README; the rest restated what any server in front of ADT is. Removed at the atlas maintainer's request on 2026-10-14.
  • arc-1description.what_it_solves, links.changelog, type_properties.tool_count.source.quote, type_properties.guardrails, changed_in, sources[].checked → the `Custom_*` extension API and the agent-plugin install in what_it_solves, the rest of the guardrails the README states, the tool-count sentence as the README writes it, the release-notes link, and 1.4.0's extension report execution as the first changed_inRe-read against the README, the npm registry record, the extensions page and the release notes on 2026-10-04, with arc-1 at 1.5.0 published 2026-10-02. Every claim the entry already carried still holds word for word, so the checked dates move and the text does not, except where the documentation says something the entry did not. Two such things: `Custom_*` tools an operator registers without forking, which the corpus already recorded on [claude-sap-kit](artifact:silvius1996-claude-sap-kit) and not on the server those tools run in, and one agent plugin that installs the server together with [ARC-1 Skills](artifact:arc-mcp-arc-1-skills). The seven guardrails added were stated when the entry was written and left out of it; they are quoted, not counted. The `run_code` facet stays absent: the extensions page gates class and report execution behind two flags and a plugin the operator writes, which is not the server stating that it runs code.
  • arc-1-extension-sampleentry → createdEntered by intake from the README, found through ARC-1's extensions page, which links it as the worked sample.
  • arc-1connects, requires, description.what_it_solves, lineage → connections to the ABAP platform over its ADT REST API and, when deployed on BTP, to an ABAP environment there and to the Destination service that forwards the user's identity; abaplint as a requirement; the abapGit and Docker links alone; the three inspirations the README namesThe connection is what the README states the server fronts (docs/concepts/design-roles-and-interfaces.md, P1), and it attaches the system's entry. The two BTP connections are stated under the README's Cloud Foundry deployment heading, so they carry that condition; the Destination service's sentence states identity forwarding and no data flow, and the ABAP environment's sentence names no protocol. The README's credits table names abaplint as what the linter uses. Two older defects go with it: a nested link that rendered its outer bracket as text, and abap-adt-api named in the same quoted sentence as the other two inspirations and carried by no edge.
  • ARC-1 Skillsentry → createdEntered from the sap-ai-mcp-servers catalogue, which lists the skills bundle as its own "ARC-1 SAP Skills & Claude Plugin" row, separate from the ARC-1 MCP server row already in the corpus as [arc-mcp-arc-1](artifact:arc-mcp-arc-1). Entered as its own artifact rather than folded into that entry, the same shape the UI5 Web Components skills take beside the UI5 Web Components library — one artifact_type per entry (hard rule 9), and the skills are a distinct install path from the server.
  • arc-1description.one_line, description.what_it_solves → corrected against the READMECorrected against the README, which documents a shared-credential mode (technical user) beside per-user identity.
  • arc1-adt-abap-mcp-extrequires[0].for, requires[0].sources[0].quote, requires[0].sources[0].checked → corrected against the READMEThe cited sentence is no longer in the README: from 0.5.1 the plugin only contributes tools to SAP's supported server; waking the dormant one was 0.3.x.
  • arc-1-lspentry → createdTriaged out of data/crawled/mcp-candidates.json, the sap-ai-mcp-servers sweep. Written from the repository's own README; the catalogue is credited as where it was found and its words are not reused.
  • arc1-adt-abap-mcp-extentry → createdTriaged out of data/crawled/mcp-candidates.json, the sap-ai-mcp-servers sweep. Written from the repository's own README; the catalogue is credited as where it was found and its words are not reused.
  • arc-1entry → createddocs/NEIGHBOURS.md §2 assigns ARC-1 the posture ingest — as an entry — and §3.1 names its author, but nothing in data/artifacts held it. Entered as the artifact the posture calls for, and it is the first provider of capability:abap-agent-tooling in the corpus.
  • arc-1provides → capability:agent-code-accessabap-agent-tooling generalised into agent-code-access: the same concern exists in CAP, and the ABAP framing keeps the local name.
  • First entry seenDiscovered via GitHub API · arc-mcp, sap-ai-mcp-servers, Added by hand.
For your agent
Ref
actor:arc-mcp
JSON
/actors/arc-mcp.json
MCP
Connecting an agent