onedevatlasBeta
Browse the atlas
AllCAPUI5ABAPBTP
SkillSkilltypeA procedure an agent follows. Indexed here, never written here.See every one →What this type is →independently maintainedindependently maintainedoriginPublished by one person or a small company outside SAP. Says nothing about quality; the readings say that.See every one →

DanKromp/skillOverflow

A collection of 41 agent skills, 37 for CAP, Fiori, BTP, UI5, ABAP, Integration Suite and HANA work and four for creating, reviewing and retiring skills, each with its own eval.

Thirty-seven skills cover CAP, Fiori, BTP, UI5, ABAP, Integration Suite and HANA work; four more cover the lifecycle of the skills themselves. A shell setup links selected categories into the global skill folders of Claude Code, GitHub Copilot, OpenCode or an agents-spec tool, leaving application repositories unchanged. Instruction files load baseline rules for every session, while slash commands and specialised agents give explicit entry points when automatic skill matching does not fire. The repository keeps one eval for each of its 41 skills in a shared eval file.

GitHub8 ★

Facts

Last pushLicenceApache-2.0

Where it sits

Lives in
CAP
Provides
Agent guidanceCAPUI5BTPABAPAgent 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.Agent evalsCAPBTPABAPAgent evalscapabilityKnowing whether an AI feature still behaves after a model, prompt or schema change. Regression testing for non-deterministic output, against fixtures that reflect real business data.
Serves
UI5ABAPBTP/BAIP

Who is behind it

DanKrompDanKrompPersonMaintains this
In the atlas: maintains 1, contributes to 1@DanKrompmaintainer

What it declares

Every artifact

Type
Skill
Origin
Independently maintained

As a skill

Distributed through
Copied into the project

For agent guidance

Installs into
Claude CodeGitHub CopilotOpenCode

For agent evals

Where the cases come from
Shipped with the package
How the output is judged
Written assertions

Tasks this contributes to1Beta

Agent evals1
Check that a coding agent still writes correct CAP codeAfter a model, a prompt or a skill set changes, the agent is given prompts whose expected answer is written down, and what it produced is checked against them.Node.js, JavaAlso in ABAPCoversone eval per shipped skill
What it does

tests/skillOverflow-evals.json states what it is for and what each case holds: a prompt a developer would write, and the assertions the answer is checked against, named one by one. The repository counts one eval for each of its 41 skills, and its CAP cases carry assertions about CDS and handler code rather than an expected file.

Remaining work

Run them, and decide what a failed assertion means. The assertions are sentences for a reader or another agent to check, so comparing an answer against them is the project's step. The cases are about this repository's skills.

All options on the Agent evals page →

Connections2

Download
DanKromp/skillOverflow connectionsDanKromp/skillOverflowPROVIDES→ provides2provides stated Pairs every listed skill with an eval in tests/skillOverflow-evals.json: a prompt a developer would write, and the assertions the agent's answer is checked against, named one by one. The assertions are sentences about the code, such as using the cuid aspect instead of a manual key. SkillOverflow — README: https://github.com/DanKromp/skillOverflow SkillOverflow — tests/skillOverflow-evals.json: https://github.com/DanKromp/skillOverflow/blob/main/tests/skillOverflow-evals.jsonAgent evalsprovides stated Packages 37 task-specific SAP instruction sets, four skill-lifecycle guides, baseline instruction files, three slash commands and seven specialised agents for four coding-agent layouts. SkillOverflow — README: https://github.com/DanKromp/skillOverflow DanKromp/skillOverflow — README, skill list, UI5: https://raw.githubusercontent.com/DanKromp/skillOverflow/main/README.md DanKromp/skillOverflow — README, skill list, Fiori elements: https://raw.githubusercontent.com/DanKromp/skillOverflow/main/README.mdAgent guidance

Install

git clone https://github.com/DanKromp/skillOverflow.git ~/skillOverflow
bash ~/skillOverflow/setup.sh

Copied from SkillOverflow — README, read . Not run by the atlas.

README

quoted from DanKromp/skillOverflow under Apache-2.0, read
DanKromp/skillOverflow/README.mdOn GitHub

SkillOverflow

License: Apache 2.0 Skills SAP

A library of AI agent skills for SAP development. Skills are markdown instruction files that AI coding agents (Claude Code, GitHub Copilot, OpenCode) load automatically when you work on relevant tasks — encoding SAP best practices so every developer gets consistent, high-quality AI assistance without having to prompt for it.

Quickstart — one time per developer, not per project

Skills are installed globally into your home folder — not inside any project. Customer repos stay clean. Every project you open gets the skills automatically.

# 1. Clone skillOverflow once into your home folder
git clone https://github.com/DanKromp/skillOverflow.git ~/skillOverflow

# 2. Run setup — choose which AI tools you use
bash ~/skillOverflow/setup.sh

The interactive setup asks which tools you use and only creates what you need:

Which AI tools do you use?
  ↑↓ move  space toggle  enter confirm  a select all

  ▶ [x] Claude Code           → ~/.claude/skills/
    [ ] GitHub Copilot        → ~/.copilot/skills/
    [ ] OpenCode              → ~/.config/opencode/skills/
    [ ] Other agents-spec     → ~/.agents/skills/

Or skip the prompt with flags:

bash ~/skillOverflow/setup.sh --claude           # Claude Code only
bash ~/skillOverflow/setup.sh --copilot          # GitHub Copilot only
bash ~/skillOverflow/setup.sh --claude --copilot # both
bash ~/skillOverflow/setup.sh --all              # all tools

Only want certain skill categories? Right after the tool picker, the interactive setup also asks which skill categories you want — all selected by default, so just deselect what you don't need:

Which skill categories do you want?
  ↑↓ move  space toggle  enter confirm  a all  n none

  ▶ [x] CAP         — Cloud Application Programming Model
    [x] Fiori       — Fiori Elements & UI development
    [x] BTP         — Business Technology Platform services
    [x] UI5         — SAPUI5 controls, binding, testing
    ...

Or skip the prompt and set the categories directly:

# Only install the CAP, Fiori, UI5 and BTP categories
bash ~/skillOverflow/setup.sh --claude --categories=cap,fiori,ui5,btp

Valid categories: cap, fiori, btp, ui5, abap, integration, hana, meta. Narrowing the set prunes categories you removed; --update keeps the categories you chose.

Then restart VS Code — skills are active in every project immediately.

Skill categories

All skills live flat under skills/<skill-slug>/ (required by the AI Skills Library) and declare their category via the metadata.category field in SKILL.md.

CategorySkillsDescription
cap22 skillsSAP Cloud Application Programming Model
fiori5 skillsSAP Fiori Elements & UI development
btp3 skillsSAP Business Technology Platform services
ui54 skillsSAPUI5 controls, binding, testing
abap1 skillABAP RESTful Application Programming Model
integration1 skillSAP Integration Suite / Cloud Integration
hana1 skillSAP HANA Cloud native / HDI development
meta4 skillsSkill lifecycle — create, review, deprecate

Total: 41 skills | 41 evals

CAP skills
SkillActivates when working on…
cds-modelingCDS entities, aspects, associations, projections
service-handlersbefore/on/after handlers, actions, functions, cds.ql
security-auth@requires, @restrict, XSUAA, IAS, row-level authorization
btp-deploymentmta.yaml, cds build, Cloud Foundry, Kyma, HDI containers
ci-cdGitHub Actions pipelines, SAP CI/CD service, deployment automation
fiori-draft@odata.draft.enabled, draft lifecycle, Fiori draft messages
cap-llm-pluginSAP AI SDK, RAG, embeddings, HANA Vector Engine, AI Core
code-reviewCAP code quality checks, anti-pattern detection, structured review
error-handlingreq.reject/warn/error, HTTP codes, i18n messages, field targets
testingcds.test, Vitest (recommended), integration tests, mocking, auth testing
messagingSAP Event Mesh, CloudEvents, emit/on patterns, outbox
multitenancy@sap/cds-mtxs, tenant provisioning, HDI per tenant
remote-servicesS/4HANA OData, BTP Connectivity, Destination, cds.connect.to
performanceN+1 queries, SELECT projections, streaming, HANA optimization
localizationlocalized elements, i18n bundles, _texts tables, CSV format
cap-pluginschange-tracking, audit-logging, attachments, notifications
observability-telemetrycds.log, @cap-js/telemetry, Cloud Logging, OpenTelemetry
cds9-migrationUpgrading to CDS 9, deprecated APIs, breaking changes
cds10-migrationUpgrading to CDS 10, Node.js 22+, bypass drafts, service results, Vitest
status-transitions@flow annotations, state machines, status workflows
cap-javaCAP Java Spring Boot, EventHandler, @Before/@On/@After, CqnService, Maven
cap-mcp-serverSAP MCP Server, UI5/Fiori MCP tooling, AI agent integration
Fiori skills
SkillActivates when working on…
fiori-elements-floorplansFloorplan selection, manifest.json, List Report, Object Page, ALP
fiori-annotationsUI.LineItem, UI.Facets, UI.FieldGroup, Criticality, ValueList
fiori-flexible-programmingBuilding blocks, extension points, controller extensions
fiori-navigationRouting, intent-based navigation, FCL, Launchpad targets
fiori-toolsApplication Generator, Page Editor, Guided Development
BTP skills
SkillActivates when working on…
btp-service-bindingsVCAP_SERVICES, cds bind, hybrid testing, Kyma ServiceBinding
btp-destinationsOAuth2ClientCredentials, OAuth2JWTBearer, principal propagation, Cloud Connector
btp-workzoneSAP Build Work Zone, HTML5 deployment, content providers, Managed App Router
UI5 skills
SkillActivates when working on…
ui5-data-bindingOData V4 model, list binding, context binding, $batch, side effects
ui5-custom-controlsControl.extend(), renderer apiVersion 2, lifecycle hooks
ui5-i18nResource model, i18n.properties, UI5Date, DateFormat, NumberFormat
ui5-testingwdi5, WebdriverIO, OPA5, page objects, QUnit, UIVeri5 migration
ABAP skills
SkillActivates when working on…
abap-rapABAP RAP, BDEF, managed/unmanaged, draft, validations, determinations, actions
Integration skills
SkillActivates when working on…
integration-suiteCloud Integration iFlows, adapters, mapping, Groovy, security material, MPL
HANA skills
SkillActivates when working on…
hana-cloud-nativeHDI containers, .hdbtable, calculation views, SQLScript, .hdbrole, hdi-deploy
Meta skills
SkillActivates when working on…
skill-creatorIdentifying the need for a new skill, full lifecycle overview
skill-contributorCreating a new skill — automates README, evals, checklist, PR draft
skill-reviewerReviewing a skill PR — structured quality checklist
skill-deprecatorRemoving outdated skills, model re-evaluation after updates

How skills work

Your prompt
    ↓
Agent reads all skill descriptions (fast — name + description only)
    ↓
Matching skill(s) load into context automatically
    ↓
Agent follows the instructions when generating code

You don't ask for skills by name — just describe your task naturally:

  • "Add a Products entity with currency" → cds-modeling
  • "Review this service handler" → code-review
  • "Set up a BTP destination with OAuth" → btp-destinations
  • "Write E2E tests for our UI5 app" → ui5-testing
  • "I want to add a new skill" → skill-contributor
If a skill doesn't load

Auto-matching is best-effort, not guaranteed — the agent decides when a skill is relevant, so occasionally one won't fire. That's by design, because SkillOverflow has three layers, and the reliable ones always have your back:

  1. Instruction files — loaded every session, guaranteed. The critical rules (cuid, managed, @requires + @restrict, no BasicAuthentication, …) always apply, even when no skill matches.
  2. Slash commands — /cap-model, /cap-review, /cap-security force-load their skill on demand. Use these when you want a specific skill for certain.
  3. Auto-loaded skills — the convenience layer that fires when your prompt matches.

So if a skill doesn't auto-load: rephrase with concrete terms (e.g. @restrict, mta.yaml, bindList), or invoke the matching slash command. You never lose the baseline rules.

Supported tools

ToolSkills folderInstruction fileSet up with
Claude Code~/.claude/skills/~/.claude/CLAUDE.md--claude
GitHub Copilot~/.copilot/skills/~/.copilot/copilot-instructions.md--copilot
OpenCode~/.config/opencode/skills/~/.config/opencode/AGENTS.md--opencode
Any agents-spec~/.agents/skills/~/.agents/AGENTS.md--agents

Skills are loaded when your prompt matches — the instruction file is loaded every session, guaranteed. Together they ensure the agent always follows SAP best practices.

Updating skills

cd ~/skillOverflow && git pull && bash setup.sh --update

--update detects which tools you already have installed and refreshes only those.

Command reference

bash setup.sh            # interactive tool picker
bash setup.sh --all      # install for all tools (Claude, Copilot, OpenCode, agents)
bash setup.sh --categories=cap,fiori,ui5  # install only skills in selected categories
bash setup.sh --update   # refresh an existing installation
bash setup.sh --status   # show what is installed and whether symlinks are healthy
bash setup.sh --list     # list all available skills by category
bash setup.sh --uninstall # remove all SkillOverflow symlinks and managed blocks
bash setup.sh --version  # print the version
bash setup.sh --help     # full usage

Stuck? Run bash setup.sh --status and share the output in the support channel — it reports broken symlinks and missing instruction blocks.

Specialized agents

Specialized sub-agents for complex SAP tasks:

AgentSkillWhat it does
cds-modelercds-modelingDesign or review a complete CDS data model from a business description
security-reviewersecurity-authFull authorization audit — @requires, @restrict, xs-security.json
handler-reviewerservice-handlersStructured code review with severity labels and concrete fixes
cap-model-explorercap-mcp-serverExplore CAP entities/services/actions using MCP before proposing changes
ui5-api-explorerui5-data-bindingDiscover UI5 APIs and provide version-aware implementation guidance
fiori-annotation-navigatorfiori-annotationsMap UX requirements to correct Fiori annotation snippets
fiori-mcp-troubleshooterfiori-toolsDiagnose Fiori MCP configuration and tool availability issues

Ask your agent: "Use the cds-modeler agent to design a data model for [domain]"

Slash commands

Three commands for the most common tasks:

CommandWhat it does
/cap-reviewFull CAP code review with severity labels
/cap-securityAuthorization security audit
/cap-modelDesign a CDS data model from a description

Learn More

First time here? Start with:

  1. ✅ Run setup.sh (all done!)
  2. 🗺️ Read the Skill Dependency Map — understand which skills work together
  3. 💡 Check Specialized Agent Commands — learn /cap-model, /cap-security, /cap-review
  4. 🧪 Run tests: bash ~/skillOverflow/tests/smoke-test.sh (verify installation)

For teams:

Pre-Commit Hooks

Prevent broken commits with automatic validation:

# One-time setup
git config core.hooksPath .githooks
chmod +x .githooks/pre-commit

The pre-commit hook runs smoke tests automatically before each commit. See .githooks/README.md for details.

Security

Report security vulnerabilities responsibly — see SECURITY.md for the disclosure process.

SkillOverflow has zero runtime dependencies — it's pure Bash and Markdown with no external packages.

Contributing

Use the skill-contributor meta skill — ask your agent:

I want to add a new skill for [topic]. Walk me through it.

All contribution steps are automated. See CONTRIBUTING.md for the full guide.

To contribute, work directly in ~/skillOverflow/:

cd ~/skillOverflow
git checkout -b skill/my-new-skill
# make changes
git push origin skill/my-new-skill
# open PR on GitHub

Versioning

# Pin to a specific version
cd ~/skillOverflow && git checkout v1.0.0

# Back to latest
cd ~/skillOverflow && git checkout main && git pull

License

Apache License 2.0 — Copyright (c) 2026 Daniel Krompaß

See LICENSE for the full text. Free to use, fork, and contribute.

Apache-2.0, read

History

Latest

v1.1.0

Changed 27

Notes
Updated — 12 skills aligned to CAP 10.1 / CAP Java 5.1 (August 2026)

All changes based on official SAP documentation: cap.cloud.sap/docs/releases

cap/cap-plugins (v1.0.0 → v1.1.0)
  • Fixed @cap-js/ai description — removed incorrect "MCP protocol exposure" claim
  • Added @cap-js/change-tracking v2 warning — database triggers, 50× perf gain, breaking schema migration required
  • Added @cap-js/mcp plugin — MCP protocol adapter that exposes CAP services to AI agents
  • Added @cap-js/print — print service integration with SAP Print Service
  • Added @cap-js/sdm — SAP Document Management Service for attachment storage
  • Added @cap-js/event-broker — SAP Cloud Application Event Hub consumer
cap/cap-mcp-server (v1.0.0 → v1.1.0)
  • Added comparison table: @sap/cds-mcp (dev tool) vs. @cap-js/mcp (protocol adapter)
cap/cap-llm-plugin (v1.0.0 → v1.1.0)
  • Fixed primary reference URL (/docs/guides/ai/ returns 404 in CDS 10)
  • Added @cap-js/ai built-in plugin section (field recommendations, AI Core Calesi integration)
  • Added note: Vector type now supported on H2, SQLite, and PostgreSQL for local development
cap/observability-telemetry (v1.0.0 → v1.1.0)
  • Added @cap-js/telemetry v2 note — OpenTelemetry SDK 2.0, adds undici HTTP instrumentation
cap/cap-java (v1.0.0 → v1.1.0)
  • Added OpenRewrite recipe for automated Java 4 → 5 migration
  • Added Built-in OData processing section (Apache Olingo removed in CAP Java 5)
  • Added Harmonized Search section (HANA search syntax default, fuzzy search on by default)
  • Added generateClasses CDS Maven Plugin parameter (up to 10× faster typed data access)
cap/cds10-migration (v1.0.0 → v1.1.0)
  • Added Node.js v26 row to version table (already tested/usable in dev)
  • Fixed Spring Boot requirement: 4.1 is now minimum, not just "recommended"
  • Added OpenRewrite recipe for automated Java migration
  • Added "Going Native" section — axios, better-sqlite3, generic-pool replaced by built-in equivalents as defaults
cap/testing (v1.0.0 → v1.1.0)
  • Added note: new CAP projects are ESM-based by default ("type": "module") in CDS 10+
cap/status-transitions (v1.0.0 → v1.1.0)
  • Fixed broken reference URL: guides/state-based-workflows → guides/services/status-flows
  • Added CAP Java Maven dependency (cds-feature-flow)
  • Added current limitations (actions disabled in draft state; CRUD/DRAFT can't be restricted)
cap/fiori-draft (v1.0.0 → v1.1.0)
  • Added cds.fiori.draft_new_action: true config option (plain POST creates active entities; draftNew action for draft creation)
cap/cds-modeling (v1.0.0 → v1.1.0)
  • Added @assert declarative constraints section (GA since CAP April 2026 / cds 9.9)
cap/remote-services (v1.0.0 → v1.1.0)
  • Added cds export GA section — publish lossless CDS API packages instead of EDMX
hana/hana-cloud-native (v1.0.0 → v1.1.0)
  • Added VECTOR_EMBEDDING native SQL function with usage example

  1. v1.0.0

    Other 31

    Notes

    Initial release. Company-wide AI agent skills library for SAP development — markdown instruction files that AI coding agents (Claude Code, GitHub Copilot, OpenCode) load automatically when working on relevant tasks.

    Skills — 41 across 8 categories
    • cap/ — cds-modeling, service-handlers, security-auth, btp-deployment, ci-cd, fiori-draft, cap-llm-plugin, code-review, error-handling, testing, messaging, multitenancy, remote-services, performance, localization, cap-plugins, cap-mcp-server, observability-telemetry, cds9-migration, cds10-migration, status-transitions, cap-java
    • fiori/ — fiori-elements-floorplans, fiori-annotations, fiori-flexible-programming, fiori-navigation, fiori-tools
    • btp/ — btp-service-bindings, btp-destinations, btp-workzone
    • ui5/ — ui5-data-binding, ui5-custom-controls, ui5-i18n, ui5-testing
    • abap/ — abap-rap (managed/unmanaged BOs, BDEF, draft, validations, determinations, actions)
    • integration/ — integration-suite (Cloud Integration iFlows, adapters, content modifier, message mapping, Groovy scripts, externalized parameters, security material, exception subprocess, MPL monitoring)
    • hana/ — hana-cloud-native (HDI containers, .hdbtable, calculation views, SQLScript procedures, .hdbrole/.hdbgrants, .hdbsynonym, Vector Engine, and how native artifacts relate to CAP)
    • meta/ — skill-creator, skill-contributor, skill-reviewer, skill-deprecator
    Specialized agents
    • cds-modeling/agents/cds-modeler.md — CDS model design and review
    • security-auth/agents/security-reviewer.md — full authorization audit
    • service-handlers/agents/handler-reviewer.md — handler code review
    Slash commands
    • commands/cap-review.md — /cap-review
    • commands/cap-security.md — /cap-security
    • commands/cap-model.md — /cap-model
    Installation
    • setup.sh — global install into ~/ (Claude → ~/.claude/skills/, Copilot → ~/.copilot/skills/, OpenCode → ~/.config/opencode/skills/, agents → ~/.agents/skills/). Customer and project repos stay clean.
    • Interactive tool selection with flags: --claude, --copilot, --opencode, --agents, --all, --update, --uninstall
    • Category (folder) selection: interactive picker (shown after the tool picker) plus the --categories=<list> flag (e.g. --categories=cap,fiori,ui5,btp) to install only the skill folders you need — narrowing prunes deselected folders and --update preserves your choice
    • Informational commands: --status (install health + broken-symlink check + staleness nudge), --list (catalog of all skills by category), --version, --help
    • Per-tool instruction files installed and kept up to date via marker-delimited blocks (<!-- SkillOverflow start/end -->) that never overwrite a developer's own instructions
    • Auto-detects WSL and updates global VS Code user settings
    Instruction files
    • instructions/CLAUDE.md, instructions/copilot-instructions.md, instructions/AGENTS.md — loaded every session so skills are always consulted
    Repository quality & governance
    • LICENSE — MIT License, Copyright 2026 Daniel Krompaß
    • SECURITY.md — vulnerability reporting policy, supported versions, zero-dependencies note
    • .github/CODEOWNERS, .github/PULL_REQUEST_TEMPLATE.md
    • .github/workflows/validate-skills.yml — validates SKILL.md presence, frontmatter, SAP doc references
    • .githooks/pre-commit — runs the smoke test and checks for backup files and credentials
    • docs/skill-map.md, commands/README.md
    • tests/smoke-test.sh, tests/SkillOverflow-evals.json (41 evals), tests/skill-test-checklist.html
    • tests/setup-test.sh — installer integration test (install / update / status / uninstall / category selection), also run in CI alongside README badge-count verification
    • All 41 skills carry version:, keywords:, and related: frontmatter
    • README.md, CONTRIBUTING.md
On GitHub

Others providing52

Agent guidance
  1. abap2UI5 for VS CodeVS Code extension to preview, activate and lint abap2UI5 appsfrom ABAP
  2. sap-cx-developer-companionAgent skill for SAP Customer Experience APIs, CDS and errorsfrom SAP Commerce
  3. sap-basis-opsRunbook skills for Basis operations at the OS and database layerfrom ABAP
  4. abap-offstack-aunitAgent skill for off-stack ABAP Unit tests and an abaplint gatefrom ABAP
  5. ARC-1 SkillsTwenty-two agent skills shipped with arc-1from ABAP
  6. sapskillsAgent skills for Clean ABAP, ADT in VS Code and guardrailsfrom ABAP
  7. SuperClaude for SAPClaude Code plugin with an ABAP MCP server and write guardsfrom ABAP
  8. sap-dm-pod2-mcp-serverMCP server for SAP Digital Manufacturing POD 2.0 extensionsfrom SAP Digital Manufacturing
  9. capire/skillsAgent skills for CAP development, each a SKILL.md with referencesfrom CAP
  10. sap-cloud-integration-pipeline-assistantAgent skill for the Pipeline Concept in SAP Cloud Integrationfrom BTP/BAIP
  11. claude-abap-devFive ABAP workflows for Claude Code on existing MCP serversfrom ABAP
  12. sap-commerce-skillAgent skill for SAP Commerce Cloud 2211-jdk21.1 and laterfrom SAP Commerce
  13. ARC-DRAWMCP server drawing BTP solution diagrams on a draw.io canvasfrom BTP/BAIP
  14. cap-mtx-scaffoldAgent skill for CAP multitenancy scaffoldingfrom CAP
  15. sap-ai-dev-toolkitMCP proxy pinning an agent's ABAP tools to one SAP systemfrom ABAP
  16. scaffold-fiori-ai-bridgeAgent skill generating a Joule Frontend Actions bridge for UI5 appsfrom UI5
  17. sapcli Claude Code PluginAgents and skills that drive sapcli instead of ADTfrom ABAP
  18. CAP GuardianCAP review skill for coding agents with gates and prohibitionsfrom CAP
  19. sap-odata-explorerRust CLI and desktop app for exploring SAP OData servicesfrom BTP/BAIP
  20. btp-solution-diagram-generatorClaude plugin writing BTP solution diagrams as .drawio filesfrom BTP/BAIP
  21. abap-skillsSeventeen agent skills for ABAP developmentfrom ABAP
  22. drawio-skill-sapGitHub Copilot skill writing BTP solution diagrams as draw.io XMLfrom BTP/BAIP
  23. sap-mdk-appsAgent skill for Mobile Development Kit projectsfrom BTP/BAIP
  24. ABAP remote filesystemVS Code extension editing ABAP systems over ADT as a filesystemfrom ABAP
  25. sap-architectureClaude Code plugin and skill drawing validated BTP diagramsfrom BTP/BAIP
  26. WerkbankLocal desktop workspace for AI conversations with SAP connectorsfrom Across stacks
  27. SAP Datasphere Plugin for ClaudeEighteen Datasphere skills over a bundled MCP serverfrom BTP/BAIP
  28. claude-abap-skillsClaude Code plugins with Clean ABAP rules and templatesfrom ABAP
  29. SAP Power for KiroSAP skills for the Kiro IDEfrom BTP/BAIP
  30. drawio-btp-diagram-skillsClaude Code skills for BTP architecture diagrams in draw.iofrom BTP/BAIP
  31. @patimweb/pi-cap-bestpracticesSAP's capire/skills packaged for the pi coding agentfrom CAP
  32. pod2-configAn agent skill that turns a described POD change into an importable [SAP Digital Manufacturing](external:sap-digital-manufacturing) POD 2.0 export, built from standard plugins and actions only.from SAP Digital Manufacturing
  33. aicore-skillsFour agent skills for SAP AI Core in SAP's AI Skills Libraryfrom BTP/BAIP
  34. SAP/automation-pilot-agent-skillsEight skills for writing and operating SAP Automation Pilot contentfrom BTP/BAIP
  35. SAP Developers MCPHosted MCP server of developers.sap.com tutorials and missionsfrom Across stacks
  36. sap-fiori-guidelinesAgent skill built from the Fiori Design Guidelinesfrom UI5
  37. sap-fiori-iosAgent skill mapping Fiori iOS UI Kit nodes in Figma to SwiftUIfrom BTP/BAIP
  38. SAP/leanix-ai-pluginsClaude Code plugin with two skills for SAP LeanIX scriptingfrom BTP/BAIP
  39. SAP-samples/html5-applications-migration-skills-for-sap-btpClaude Code plugin with skills migrating HTML5 apps to appFrontfrom BTP/BAIP
  40. SAP-samples/joule-a2a-agent-toolkitSkills for building an A2A agent for Joule and deploying it to CFfrom BTP/BAIP
  41. SAP-samples/sap-devs-cliCLI adding SAP developer knowledge and MCP servers to AI coding toolsfrom BTP/BAIP
  42. sap-tutorials-contentAgent skill mapping the developers.sap.com content APIfrom BTP/BAIP
  43. SAP/ui-theme-designer-plugins-for-coding-agentsAgent plugin on the SAP Design System and theming parametersfrom UI5
  44. secondsky/sap-skillsForty SAP development plugins for coding agentsfrom BTP/BAIP
  45. claude-sap-kitClaude Code plugin adding Custom_* tools to arc-1from ABAP
  46. UI5/plugins-coding-agentsUI5 plugins for coding agentsfrom UI5
  47. UI5 Web Components agent skillsTwo agent skills for accessibility and styling of web componentsfrom UI5
  48. rap-skillsRAP skills delivered branch-per-target for each deploymentfrom ABAP
  49. KlystronMCP server in ABAP inside the application server, with agent skillsfrom ABAP
Agent evals
  1. agentic-abap-evalsABAP navigation tasks and a grader for coding agentsfrom ABAP
  2. capire/skillsAgent skills for CAP development, each a SKILL.md with referencesfrom CAP
  3. LLM-Benchmark-ABAP-Code-GenerationBenchmark of LLM ABAP code generation against a real SAP systemfrom ABAP
  4. @saptarishi/cds-plugin-llmCAP service kind putting eleven model providers behind one APIfrom CAP

Details

Sources3
SAP AI Skills Librarydiscovered here, skills.cloud.sapchecked
github.com/DanKromp/skillOverflowtree, pushes, contributors, CIchecked
README.mdquoted, and what the entry declareschecked

Sources are graph entities: the same page may back several entries, and an unreachable source flags every row that rests on it.

Compatibility and provenance
discovered viaSAP AI Skills Library · skills.cloud.sap
first seen in the atlas
ecosystem and domaincap, ai-and-agents
History in the atlas5
  1. Pairs every listed skill with an eval in tests/skillOverflow-evals.json: a prompt a developer would write, and the assertions the agent's answer is checked against, named one by one. The assertions are sentences about the code, such as using the cuid aspect instead of a manual key.
  2. Claude Code, GitHub Copilot, OpenCode
  3. Two rows of the README's skill list, one UI5 and one Fiori elements
  4. Discovered through the SAP AI Skills Library and represented once at repository level. The registry listed 37 contained skills when checked; the canonical repository contained and documented 41, so the repository count is used for the artifact while both snapshots remain in the crawl record.
  5. First seen.
For your agent
Ref
artifact:dankromp-skilloverflow
JSON
/artifacts/dankromp-skilloverflow.json
MCP
Connecting an agent