Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

Mcp Tool Scout — measured AI agent skill | DecimalAI

stijnman/mcp-tool-scout

SkillScore91Maintained

Discovers MCP servers and reads tool schemas before calling MCP tools. Use b efore MCP invocations or when user says what MCP tools, check MCP schema. Re ad-only filesystem inspection of MCP descriptors. Triggers: mcp tools, MCP s chema,discover MCP. Use when the user needs this capability.

by stijnman

Links this skill to your agents and keeps it current with the author’s updates.

Automatic

AI Engineer pack

decimalai skills pull stijnman-mcp-tool-scout

Get this skill — curl · raw file

curl -fsSL https://app.decimal.ai/s/stijnman-mcp-tool-scout/SKILL.md

Raw file: /s/stijnman-mcp-tool-scout/SKILL.md ↗ · machine-readable summary at /s/stijnman-mcp-tool-scout.json ↗

In productionGlobal

how this skill behaves on real traffic · last 30 days

No production data for this skill yet — usage numbers appear here once agents start running it.

Anonymized aggregate across every org using this skill — proof from real production use, not a demo.

Does this skill help? · Benchmark: with vs. withoutsign in to run this on your agents

22 test cases from the publisher’s eval.yaml · each run twice on gemini-3.6-flash — with the skill injected, then without · last run 7d ago

Yes — +32% lift-23% tokens0% turns91% of whole cases pass with skill (20/22)

Test case	Without → With	Effect	Δ tokens	Δ turns
case-03	✗→✓	▲ Improved	21%	0%
case-04	✗→✓	▲ Improved	-13%	0%
case-09	✗→✓	▲ Improved	38%	0%
case-10	✗→✓	▲ Improved	-40%	0%
case-11	✗→✓	▲ Improved	-57%	0%

Showing 5 of 22 cases

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 8/1/2026

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

stijnman / grok-custom-skills

.grok/skills/mcp-tool-scout/SKILL.md

View on GitHub ↗

License MITImported 8/1/2026Version v1

Skill contents

SKILL.md

Mcp Tool Scout

When to Use

User says mcp tools or task matches this capability
User says MCP schema or task matches this capability
User says discover MCP or task matches this capability
User says which MCP tool or task matches this capability

Workflow

List MCP server folders (e.g. mcps/<server>/tools/.json).
Match user task keywords to available tool names and descriptions.
Read full inputSchema for chosen tool before any CallMcpTool.
Report missing auth or server instructions; do not guess parameters.
Prefer least-privilege tool selection.

Integrations

tool-discovery-engine
defensive-mcp-audit
hitl-approver

Error Handling

| Failure | Response | |---------|----------| | Schema missing | Stop; do not call tool without reading descriptor. | | Auth required | Ask user to complete auth out-of-band; never embed secrets in skill. |

Gotchas

Mandatory schema read prevents parameter injection mistakes.

Safety & Ethics (Publication-Ready)

This skill is designed for public distribution. Constraints:

Read-only discovery; no MCP calls without user task context.
Never log or publish auth tokens from MCP configs.
Tool paths use forward slashes; no Windows credential stores.

Prohibited actions

No unauthorized access, malware, or harmful automation
No silent exfiltration of data, credentials, or telemetry
No destructive system changes without hitl-approver
No publication of user PII or environment secrets in outputs

Example

Input: User request matching triggers above. Output: Structured result per workflow; local artifacts only unless user opts in.

Want to see if stijnman/mcp-tool-scout would activate on your phrasing? Try it before you install — no LLM call, no signup.

Other measured skills in the registry, with their headline benchmark lift.