CtrlK
BlogDocsLog inGet started
Tessl Logo

soccer-agent-toolbelt

Soccer analytics agent toolbelt. Gives Claude Code direct access to the Oracle-backed match data, ML predictions, and three-tier memory. Use when answering questions about football matches, building on the soccer agent, or exploring the World Cup dataset.

65

Quality

78%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./workshops/soccer-analytics-agent/.claude/skills/soccer-agent-toolbelt/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

75%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A highly actionable, well-structured toolbelt reference with copy-paste commands and concrete Oracle-specific gotchas. Its main weakness is redundancy — the hybrid-first and observability rules are each restated multiple times — which costs token efficiency.

Suggestions

State the hybrid-first rule once in the "Hybrid-first rule" section and have the `hybrid_retrieve` tool entry and the Pitfalls entry cross-reference it instead of repeating it.

Collapse the "LangGraph OracleDB observability rule" and the "Observe the loop" pitfall into a single section to remove the duplicated observability explanation.

Add a verification checkpoint to the "Adding your own tool" workflow (e.g., run the dispatcher against the new tool and confirm a JSON result) so the sequence includes a feedback loop.

DimensionReasoningScore

Conciseness

Mostly dense and earns its place, but the hybrid-first rule is restated three times (the "Hybrid-first rule" section, the `hybrid_retrieve` tool description, and the "Hybrid-first default" pitfall) and observability is explained twice, which is unnecessary re-explanation that could be tightened; this lands at the mostly-efficient-but-could-tighten anchor rather than the lean 4.

3 / 5

Actionability

Guidance is fully executable: exact `uv run python ... run_tool.py` dispatch commands with complete JSON arg examples, a copy-paste curl observability snippet, a Python heredoc, and concrete SQL — covering the common cases copy-paste ready.

5 / 5

Workflow Clarity

Sequences are clear with a real feedback loop in the observability section ("If this returns no rows... run init_memory.py and verify.py; the verifier must report..."), and adding a tool is a clean 3-step list; it stays at 4 because the "Adding your own tool" flow has no validation checkpoint before the shared dispatcher picks the tool up.

4 / 5

Progressive Disclosure

No bundle files exist (references/scripts/assets absent) and everything is inline, but the body is well-organized into clear sections (Tools, How to invoke, Adding your own tool, Pitfalls) with signaled navigation; the substantial Pitfalls block is appropriately scoped inline for a workshop gotchas reference, giving good-but-not-perfect structure.

4 / 5

Total

16

/

20

Passed

Description

82%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong description that clearly states both capability and trigger context with a distinctive, low-conflict niche. Its only weakness is specificity: it enumerates data sources rather than concrete actions a user would invoke.

Suggestions

Replace "Gives Claude Code direct access to" with concrete verbs, e.g. "Query match data, run ML match predictions, and search three-tier memory".

Add a couple of natural trigger synonyms such as "fixtures", "results", or "team ratings" to broaden keyword coverage.

DimensionReasoningScore

Specificity

Names the domain and three concrete capabilities ("Oracle-backed match data, ML predictions, and three-tier memory") but the only action stated is "Gives Claude Code direct access to" — it lists objects rather than multiple distinct actions, so it stops at the 1-2 concrete actions anchor rather than the several-actions anchor at 4.

3 / 5

Completeness

It explicitly answers "what" ("Gives Claude Code direct access to the Oracle-backed match data, ML predictions, and three-tier memory") and "when" with concrete triggers ("Use when answering questions about football matches, building on the soccer agent, or exploring the World Cup dataset"), matching the explicit what-and-when anchor.

5 / 5

Trigger Term Quality

Natural phrases a user would say are present ("questions about football matches", "building on the soccer agent", "exploring the World Cup dataset") with both "football" and "soccer" covered; a few common variations (e.g., specific team or fixture terms) are missing, so it is just below the comprehensive 5.

4 / 5

Distinctiveness Conflict Risk

The soccer/World-Cup/Oracle niche is highly specific with distinct triggers, making accidental triggering of an unrelated skill unlikely; it sits clearly at the minimal-conflict anchor rather than the minor-overlap 4.

5 / 5

Total

17

/

20

Passed

Validation

93%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

referenced_paths_exist

Referenced path issues: 4 missing

Warning

Total

15

/

16

Passed

Repository
oracle-devrel/oracle-ai-developer-hub
Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.