exekiel179/spss-mcp/spss-mcp-guard

Claude Code and agent skill by exekiel179/spss-mcp

Trust tier 3
  • community
Source check pending

Checking that the SKILL.md file is still in the repository.

Make IBM SPSS Statistics MCP execution more reliable by preventing common MCP/SPSS failures before they happen and diagnosing them when they do. Use when the user wants to avoid SPSS MCP timeouts, syntax incompatibilities, PROXSCAL/MDS errors, invalid subcommands, stale environment configuration, or flaky batch execution; when a statistical analysis should be executed defensively with smoke tests first; or when Claude should probe with minimal syntax before attempting a heavier SPSS run.

Install

npx vskill@latest install exekiel179/spss-mcp/spss-mcp-guard

SKILL.md

The same SKILL.md was published earlier as opalus12138/spss-mac-mcp/spss-mcp-guard.

SPSS MCP Guard

Use this skill to make SPSS MCP execution robust, especially for uncommon procedures, long-running analyses, and syntax that may vary by SPSS version.

Workflow

  1. Check capability first
  • Call spss_check_status.
  • If SPSS batch is unavailable, stop and explain that only file-mode tools are safe.
  1. Read the data before writing syntax
  • Use spss_file_summary or spss_read_metadata first.
  • Never guess variable names or data shape.
  1. Prefer the safest execution path
  • Use a dedicated MCP tool when one exists and is known-good.
  • Use spss_run_syntax for uncommon methods, but only after a smoke test.
  1. Smoke test before heavy analysis
  • First run a minimal procedure such as:
  • DESCRIPTIVES
  • FREQUENCIES
  • DISPLAY DICTIONARY
  • Only attempt heavier methods after the smoke test succeeds.
  1. Escalate analysis gradually
  • Start with the smallest sample or simplest valid syntax.
  • Add options one subcommand at a time.
  • For MDS/PROXSCAL-style work, first validate the command family and legal keywords by interpreting SPSS warnings.
  1. Treat SPSS warnings as syntax guidance
  • If SPSS says a keyword is unrecognized, rewrite syntax to match the reported valid keywords.
  • Do not keep retrying the same syntax.
  1. Distinguish timeout from syntax failure
  • A long wait is not always a true performance issue.
  • If possible, probe the same command through a minimal direct execution path before concluding it is only a timeout issue.

Guardrails

  • Do not jump directly to complex procedures on first attempt.
  • Do not assume old SPSS syntax examples match the installed version.
  • Do not assume .env settings are loaded; verify behavior if timeout values look stale.
  • Do not treat success=True as “analysis valid” without reading warning blocks.
  • For uncommon procedures, keep the first successful run minimal, then add output options later.

Decision Tree

  1. Execution fails immediately?
  • Check spss_check_status.
  • Verify data path and variables with file tools.
  1. Execution times out?
  • Re-run a minimal baseline procedure on the same file.
  • If baseline succeeds, suspect syntax/procedure complexity rather than global SPSS failure.
  • Inspect timeout configuration behavior before increasing complexity.
  1. SPSS returns warnings/errors but output exists?
  • Read the warning text carefully.
  • Use it to rewrite subcommands.
  • Especially common for PRINT, MODEL, and SHAPE subcommands in advanced procedures.
  1. Procedure is unfamiliar or cold-path?
  • Use spss_run_syntax.
  • Start with minimal valid syntax.
  • Add optional subcommands only after the base command works.

Known reliable patterns

  • For a baseline smoke test:
spss
DESCRIPTIVES VARIABLES=x1 x2 x3
/STATISTICS=MEAN STDDEV MIN MAX.
  • For defensive PROXSCAL probing, prefer minimal legal syntax first. Do not assume keywords like SYMMETRIC, EUCLID, or CONFIGURATION are valid in the installed SPSS syntax variant without testing.

References

  • Read references/failure-patterns.md when diagnosing timeouts, invalid keywords, stale .env behavior, or advanced procedure incompatibilities.
  • Reuse spss-analyst for general SPSS workflow and syntax conventions; this skill is specifically for failure prevention and recovery.

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

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
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/exekiel179/spss-mcp/spss-mcp-guard/badge)](https://verified-skill.com/skills/exekiel179/spss-mcp/spss-mcp-guard)

Something wrong with this skill or its page? Tell us.