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.
npx skills add marcellourbani/vscode_abap_remote_fs --skill run-scriptsgit 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/skills/marcellourbani/vscode_abap_remote_fs/run-scripts)<a href="https://agentmods.dev/skills/marcellourbani/vscode_abap_remote_fs/run-scripts"><img src="https://agentmods.dev/badge/skills/marcellourbani/vscode_abap_remote_fs/run-scripts/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/skills/marcellourbani/vscode_abap_remote_fs/run-scripts"><img src="https://agentmods.dev/badge/skills/marcellourbani/vscode_abap_remote_fs/run-scripts.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 10 findings, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium MCP Rug Pull · line 49 npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.Fix: Pin the version: npx @scope/[email protected]
- medium MCP Rug Pull · line 197 npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.Fix: Pin the version: npx @scope/[email protected]
- medium Prompt Injection · line 175 Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.Fix: Remove the large whitespace padding (blank-line blocks or long space runs) and review any content hidden below or to the right of it. Keep skill files compact and reviewable so no instructions can be
- medium Prompt Injection · line 177 Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.Fix: Remove the large whitespace padding (blank-line blocks or long space runs) and review any content hidden below or to the right of it. Keep skill files compact and reviewable so no instructions can be
- medium Prompt Injection · line 178 Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.Fix: Remove the large whitespace padding (blank-line blocks or long space runs) and review any content hidden below or to the right of it. Keep skill files compact and reviewable so no instructions can be
- medium Prompt Injection · line 188 Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.Fix: Remove the large whitespace padding (blank-line blocks or long space runs) and review any content hidden below or to the right of it. Keep skill files compact and reviewable so no instructions can be
- medium Prompt Injection · line 180 Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.Fix: Remove the large whitespace padding (blank-line blocks or long space runs) and review any content hidden below or to the right of it. Keep skill files compact and reviewable so no instructions can be
- medium Prompt Injection · line 183 Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.Fix: Remove the large whitespace padding (blank-line blocks or long space runs) and review any content hidden below or to the right of it. Keep skill files compact and reviewable so no instructions can be
- medium Prompt Injection · line 189 Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.Fix: Remove the large whitespace padding (blank-line blocks or long space runs) and review any content hidden below or to the right of it. Keep skill files compact and reviewable so no instructions can be
- medium Prompt Injection · line 190 Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.Fix: Remove the large whitespace padding (blank-line blocks or long space runs) and review any content hidden below or to the right of it. Keep skill files compact and reviewable so no instructions can be
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 | $0.00081 | $0.06305 |
| Opus 5.5 | $0.00032 | $0.02522 |
| Sonnet 5.5 | $0.00016 | $0.01261 |
| Haiku 4.5 | $0.00008 | $0.00630 |
Grade A, and why
run-scripts 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 1mo 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 — 229 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Run Scripts — Phase 7 (of 7)
If selected cases are linked stages across SAP connections, load
multi-system-workflows alongside this skill and execute separate connection-specific
tool calls in stage order. Never put dependent stages in one parallel invocation.
Phase order: analyze-and-plan (1) → explore-ui (2) → design-cases (3) → define-data (4) → prepare-data (5) → build-scripts (6) → run-scripts (7).
For bounded, self-contained support work, use sap-task-helper with explicit inputs, allowed writes, and an output contract.
Tool availability (read this if a tool seems missing)
Naming: when these docs say call X, X is a tool (you invoke it and get a result); delegate to / invoke X is an agent (a subagent you launch); load / follow X is a skill (a procedure you read). A name without a verb: see the overview's "Skills, tools, and agents" list.
The editor may hide tools until searched for, and this is exactly where it bites: abapfs_run_playwright_tests is frequently NOT in the default active toolset, and smaller models fail to find it and waste turns trying run_in_terminal/runTests instead. Before Step 0, make sure abapfs_run_playwright_tests, abapfs_get_test_folder, abapfs_get_connected_systems, abapfs_check_test_data, and abapfs_build_evidence_report are available; if any is missing, search your available tools for it by name. Never use VS Code's generic runTests tool for SAP specs — it expects Vitest/Jest and returns "No tests found" for Playwright specs. If abapfs_run_playwright_tests cannot be found after searching, tell the user rather than improvising.
The abapfs_run_playwright_tests prerequisite gate
abapfs_run_playwright_tests is a GATED tool: it refuses to run unless you pass the mandatory prerequisiteConfirmation field. Only pass it after you have completed Step 0 (artifacts present AND abapfs_check_test_data clean for the selected cases). The exact text to pass is: I verified all upstream phase gates and test data readiness for this program. It is valid only once that gate is genuinely complete; the run is rejected otherwise.
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.
- 1mo ago First seen · 229 lines · 81 tokens per session scan A b838fab04cc0
run-scripts is a skill published in the GitHub repository marcellourbani/vscode_abap_remote_fs (402 stars, last pushed 3d ago), licensed MIT. It adds 81 tokens to every session and 6,305 once invoked, about $0.0003 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 2026-08-30.
Other skills, from other repositories
sap-sac-test-automation
SAP Analytics Cloud (SAC) automated testing skill for designing capability-gated browser discovery and deterministic Playwright test suites for SAC stories, dashboards, reports, planning workflows, comments, permissions, visual regression, and reusable QA automation. This skill should be used when building SAC…
sap-fiori-testing
SAP Fiori/UI5 testing, accessibility, and modern UI skill. Use when writing wdi5 E2E tests, OPA5 integration tests, implementing WCAG 2.1 accessibility, using UI5 Web Components, or migrating UI5 to TypeScript. If the user mentions wdi5, OPA5, Fiori test, UI5 accessibility, WCAG, or UI5 TypeScript migration, use this…
sap-tcd-chain
Drives a complete headless O2C business-document chain — sales order → outbound delivery → goods issue → billing — over RFC BAPIs, so regression testing gets whole document flows, not isolated documents. Each step is created via its SAP write BAPI, committed (WAIT='X'), and VBFA-verified before the next step (VBTYPN…
sap-test-replay
Turns a recorded business scenario into a re-runnable pass/fail GUI regression test — the eCATT/CBTA equivalent the suite lacks — so a regression cycle stops meaning "re-test transactions by hand and paste screenshots into Word". A scaffold-recorded linear scenario (control IDs + screen identity + popup transitions…
sap-user-guide
Turns a recorded transaction walkthrough into an end-user training guide + UAT script — the unfunded rollout mandate everyone hand-makes from screenshots pasted into Word — composed from ground truth the suite already owns: /sap-gui-probe drives/records the transaction step by step with full screen identity, DDIC…
verify-omnigent
Drive Omnigent the way a user does and prove a behavior with recorded evidence, using an isolated server, runner, and mock model plus a feature map of every user entry point. Load before reproducing a user-facing bug, before claiming a fix works, or when reviewing whether a change covered every surface (session vs.…