ditrixnew/edt-mcp/edt-mcp-project-external-objects
Claude Code and agent skill by ditrixnew/edt-mcp
- community
Checking that the SKILL.md file is still in the repository.
Work with external data processor/report projects through EDT-MCP: metadata, forms, builds, and one-run debug launch. Not for prebuilt files outside an EDT project.
Install
npx vskill@latest install ditrixnew/edt-mcp/edt-mcp-project-external-objectsSKILL.md
EDT-MCP external objects
Purpose and trigger
Use this skill for metadata, code, forms, builds, or a bounded debug run of an external data processor/report that exists in an EDT external-object project.
Operating rule
Read and apply the common operating rules before this workflow.
Task boundary
Select the exact external-object project and qualified ExternalDataProcessor.<Name> or ExternalReport.<Name> target. Do not substitute its linked base configuration, a normal configuration object, or a prebuilt .epf/.erf outside an EDT project.
Primary workflow
- Resolve the project, object, linked base application, and runtime target with
list_projects, get_applications, and list_configurations.
- Route metadata structure to
edt-mcp-project-metadata, BSL changes to
edt-mcp-project-local-fix, and managed-form work to edt-mcp-project-forms; re-read and validate the exact owning object after a mutation.
- For a build, resolve the qualified target and compare its simple name across
both external-object kinds. objectName is a simple-name selector; stop on a processor/report collision unless building both is explicitly authorized. Before building, obtain explicit update/restructure authority for the associated build infobase, or confirm that it is disposable; the platform dumper may prepare and update that infobase even for an ordinary build. Call build_external_objects into a fresh empty staging directory, omitting objectName only for an authorized build-all. Set recordBuildTime=false unless changing the source object's Comment was requested. Verify the result and staged artifact before promotion; preserve an existing artifact and authorize its exact replacement before any in-place build or promotion.
- For a debug run, complete target, launch-policy, disclosure, credential, and
debug_status preflight first, using set_infobase_credentials only when authorized. If a matching live client exists, use it only as an explicitly authorized target; stop when the external object requires a fresh launch. Otherwise call set_breakpoint before launching the resolved external-object target according to the current launch help/schema. Retain task-owned IDs and verify that the returned target identity matches the resolved target. If a preflight race returns alreadyRunning=true, refresh the uniquely identified target, resume only a task-caused suspension, remove task-owned temporary state, and stop. Terminate and relaunch only with explicit user authorization.
- After launch, settle the task-owned launch outcome under current tool
help/schema and a bounded caller-approved deadline before cleanup, an absence claim, or completion. If the deadline expires, report the launch as in-flight or unknown with its outstanding task-owned cleanup obligation; do not assume a later client cannot appear. Once settled, call wait_for_break and inspect with get_variables only when the current help proves the intended debug target is unambiguous. Otherwise remove only task-owned temporary state and stop.
- Finish with
resumefor any task-suspended execution, calling
remove_breakpoint, and using terminate_launch only for a uniquely identified, task-owned launch whose termination is authorized.
Authority rule
Building, credentials, infobase update/restructure, external-change handling, launch/restart, artifact replacement, data disclosure, and termination require the authority applicable to their exact targets and effects.
Stop rule
Stop on wrong project kind, ambiguous object/runtime/debug target, unsupported write, missing prerequisite, unavailable credentials, or unapproved side effects. Never operate on an unrelated active launch to make the route work.
Completion signal
Report the exact project/object, confirmed changes or build artifacts, bounded runtime evidence when requested, partial or unproved claims, and cleanup of all task-owned temporary state.
Security
- Security scanPassed, 90 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingagent/skills/edt-mcp-project-external-objects/SKILL.md
Popularity
- 7-day trend score
- 0
Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/ditrixnew/edt-mcp/edt-mcp-project-external-objects)