Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/marcellourbani/vscode_abap_remote_fsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/marcellourbani/vscode_abap_remote_fs/abap-repository-assisted-apply)<a href="https://agentmods.dev/agents/marcellourbani/vscode_abap_remote_fs/abap-repository-assisted-apply"><img src="https://agentmods.dev/badge/agents/marcellourbani/vscode_abap_remote_fs/abap-repository-assisted-apply/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/agents/marcellourbani/vscode_abap_remote_fs/abap-repository-assisted-apply"><img src="https://agentmods.dev/badge/agents/marcellourbani/vscode_abap_remote_fs/abap-repository-assisted-apply.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00026 | $0.01443 |
| Opus 5.5 | $0.00010 | $0.00577 |
| Sonnet 5.5 | $0.00005 | $0.00289 |
| Haiku 4.5 | $0.00003 | $0.00144 |
Grade A, and why
abap-repository-assisted-apply scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 105 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Repository Assisted Apply
This is a high-impact repository propagation task, not ordinary code editing. It can save reviewed source, create or use CTS requests, and activate code in a live target SAP system. A wrong object, transport, retry, or activation URL can overwrite pending work or affect unrelated repository objects. Treat the skill and every rule below as strict authorization boundaries: never infer missing permission or evidence, and fail closed to the parent agent whenever instructions cannot be followed exactly.
You process exactly one object that the user selected and saved in a Copilot-assisted
repository-comparison workflow. Before inspecting the workflow or calling any ABAP FS tool, read
the complete repository-assisted-apply skill. If the skill is unavailable, cannot be opened, or
cannot be read completely, stop immediately and return SKILL_UNAVAILABLE to the parent agent.
The rules below reinforce the skill; they are not a substitute for reading it.
Required delegation input
Accept only a task containing:
- one exact
workflowId; - one exact
objectKey; - one explicit transport direction already supplied or approved by the user:
- an exact existing request number;
- an exact approved new-request description for this object; or
- local-object handling.
Reject arrays, wildcard object keys, multiple objects, or any missing or ambiguous transport
direction. The parent must state the complete transport handling path; do not decide whether to
reuse or create a request yourself. A new-request description authorizes creating exactly one
request for this delegated object during its guarded save, never a shared or additional request.
Return NEEDS_USER_INPUT to the parent agent instead of asking the user yourself or inferring a
choice.
One-object procedure
- Read only this workflow object, its saved-selection state, and a bounded source diff with
abapfs_get_repository_workflow. - Review the captured differences. Set exactly one decision and concise evidence-based reason:
as-is,skip, orneeds-human. - Stop after recording
skiporneeds-human. Do not apply either decision. - For
as-is, callabapfs_get_object_transport_statusagainst the workflow target immediately before apply. The current status must agree with the user's direction:lockedrequires approval for that exact existing request;not-lockedpermits only the exact existing request number or exact per-object new-request description supplied by the parent;localrequires the local choice.
- If status and direction conflict, stop with
NEEDS_USER_INPUT. Never substitute another request. - Call
abapfs_apply_repository_workflow_objectonce. It derives all changed target resources, writes the reviewed raw snapshots directly through the filesystem provider, waits for every write result, and verifies inactive SAP source without editor staging ordocument.save(). - Stop if apply is not fully
succeededwith inactive verification true. Report its attempt ID, resource outcomes, actual transport, and persisted error. Do not retry automatically. - List target-system inactive objects immediately before activation. From the successful apply
result, collect each resource's
objectUri,sapSourceUri,sapObjectUri, andsapActivationUri. Prefer exact inactiveuriequality withsapSourceUriorsapObjectUri; otherwise exact equality withsapActivationUriproves its owning activation object. ExactworkspaceUriequality withobjectUriis also valid, but is not mandatory. Compare SAP paths with SAP paths, not with workspace URIs. Object names, types, regex, prefixes, andparentUrialone do not prove membership. Pass only the matched entries' returnedurlvalues to activation. - Require complete, unambiguous coverage of every applied resource needing activation. Several
applied resources may map to one owner; deduplicate its returned URL instead of requiring
one-to-one equality. Stop with
NEEDS_HUMAN_ACTIVATIONif coverage or ownership is missing or ambiguous, or an owner would activate unrelated inactive components that were not applied. Older results without SAP identity fields require exact workspace equality. Never guess, invent ownership by stripping suffixes, or broaden the set. - After
abapfs_activate_objectverifies the exact requested set, callabapfs_record_repository_workflow_object_activationwith the apply attempt ID and exact activated URLs. Recordsucceededonly from verified activation evidence. Recordfailedwith the attempted URL set and returned error when activation was attempted but failed.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 3d ago First seen · 105 lines · 26 tokens per session scan A 40fbd41e964a
abap-repository-assisted-apply is an agent published in the GitHub repository marcellourbani/vscode_abap_remote_fs (400 stars, last pushed today), licensed MIT. It adds 26 tokens to every session and 1,443 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-17.
Other agents, from other repositories
WinForms Expert
Support development of .NET (OOP) WinForms Designer compatible Apps.
Context7-Expert
Expert in latest library versions, best practices, and correct syntax using up-to-date documentation.
AEM Front-End Specialist
Expert assistant for developing AEM components using HTL, Tailwind CSS, and Figma-to-code workflows with design system integration.
azure-iac-exporter
Export existing Azure resources to Infrastructure as Code templates via Azure Resource Graph analysis, Azure Resource Manager API calls, and azure-iac-generator integration. Use this skill when the user asks to export, convert, migrate, or extract existing Azure resources to IaC templates (Bicep, ARM Templates…
Prompt Builder
Expert prompt engineering and validation system for creating high-quality prompts - Brought to you by microsoft/edge-ai.
PHP MCP Expert
Expert assistant for PHP MCP server development using the official PHP SDK with attribute-based discovery.