CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

moai-foundation-cc

Canonical Claude Code authoring kit covering Skills, sub-agents, plugins, slash commands, hooks, memory, settings, sandboxing, headless mode, and advanced agent patterns. Use when creating Claude Code extensions or configuring Claude Code features.

Quality

79%

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 ./.claude/skills/moai-foundation-cc/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

67%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 well-structured, largely actionable authoring reference with a clear documentation index and a rare explicit verification checklist. Its main weaknesses are token efficiency — digressive MoAI-internal policy, a version-history section, and date/version asides scattered mid-text — and creation workflows whose validation lives only in the closing checklist rather than inline.

Suggestions

Move the MoAI-internal nesting policy detail (manager-lead fan-out seam, DEPTH_SEAL_VIOLATION sentinel) and the /agents removal changelog deltas out of the Sub-agents quick reference and into reference/advanced-agent-patterns.md or a dedicated changelog section, keeping the body to the portable facts (default depth, env var, disallowedTools).

Replace the version/date asides (CC 2.1.197/2.1.217/2.1.198, 2026-07-03) with a single 'Version-sensitive notes' or deprecated-section block so time-sensitive detail stops competing with evergreen guidance, and drop the Version History section.

Add inline validation steps to each creation workflow (e.g., 'run /plugin validate . after writing plugin.json'; 'reload and confirm the skill appears in /skills output') instead of relying solely on the closing Verification checklist, and include a concrete plugin.json example to match the specificity of the other sections.

DimensionReasoningScore

Conciseness

The body is mostly terse quick-reference bullets, but it contains several padded or digressive passages: the 150+ word Sub-agents Quick Reference paragraph wanders into MoAI-internal policy ("manager-lead carries it as the sole depth-1 fan-out seam... sentinel DEPTH_SEAL_VIOLATION"), a Version History section, and scattered version/date-sensitive asides ("CC 2.1.197", "CC 2.1.217", "2026-07-03 doc lag") that the rubric says should penalize conciseness since they are not placed in a deprecated/old-patterns section. This matches anchor 3: mostly efficient but includes unnecessary explanation that could be tightened — not anchor 2, because padding is confined to a few sections rather than being pervasive.

3 / 5

Actionability

Concrete, executable guidance dominates: real paths (~/.claude/skills/, .claude/agents/), actual commands ("/plugin install owner/repo", "/plugin validate .", "moai ast-grep"), flags (--allowedTools, -p, --dangerously-skip-permissions), and specific search examples ("sg -p 'class $X extends Service' --lang python", "Grep(pattern=\"func \", path=\"internal/core/\")"). Minor gaps keep it at anchor 4: the plugin manifest section describes the JSON shape ("name, description, version as 1.0.0, author object") without an actual example, and a few Quick Reference lines are pure summary. Not anchor 5, because several creation workflows describe fields rather than giving copy-paste-ready file templates.

4 / 5

Workflow Clarity

Multi-step processes are clearly sequenced (Explore-Plan-Code-Commit's four phases, the three-level progressive disclosure flow, sub-agent file format followed by critical rules), and the body closes with an explicit Verification checklist ("settings.json is valid JSON", "Every hook in settings.json has a matching script file") plus Red Flags — real validation checkpoints. It sits at anchor 4 rather than 5 because the individual creation workflows (skill, sub-agent, plugin) lack inline validate-and-retry loops; validation is consolidated in a global checklist at the end instead of appearing at each step where errors occur.

4 / 5

Progressive Disclosure

The Documentation Index clearly signals 14 one-level-deep reference files with one-line descriptions, and the body stays at overview altitude with a "Resources" pointer to the reference directory. This matches anchor 4 (good structure, references mostly clear, minor organization gaps) rather than anchor 5: no bundle files were present to verify the referenced paths exist, the paths use "reference/" while the convention for skill bundles is a "references/" directory, and the Quick Reference plus Common Rationalizations/Red Flags/Verification sections inline content that overlaps what the indexed reference files should carry.

4 / 5

Total

15

/

20

Passed

Description

83%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: it names a clear niche, enumerates the concrete feature areas it covers, and includes an explicit 'Use when...' trigger clause with actionable phrasing. Its only weaknesses are verb-level genericity ("covering") and a slightly broad "configuring Claude Code features" trigger that could overlap sibling configuration skills.

DimensionReasoningScore

Specificity

The description enumerates a comprehensive list of concrete feature areas ("Skills, sub-agents, plugins, slash commands, hooks, memory, settings, sandboxing, headless mode, and advanced agent patterns"), but the verbs are limited to "covering" / "authoring kit" rather than distinct concrete actions. It lists several specific targets with minor gaps in action coverage, matching the anchor at 4 rather than 5, which requires multiple specific concrete actions.

4 / 5

Completeness

It explicitly answers both questions: the "what" is "Canonical Claude Code authoring kit covering [12 named areas]" and the "when" is a concrete trigger clause — "Use when creating Claude Code extensions or configuring Claude Code features". This mirrors the anchor-5 pattern (clear what + explicit 'Use when' with concrete trigger phrases); the when-clause is present and specific, so the cap of 3 for a missing trigger does not apply.

5 / 5

Trigger Term Quality

Natural keywords users would say are present and specific: "Skills", "sub-agents", "plugins", "slash commands", "hooks", "memory", "settings", "sandboxing", "headless mode", plus "creating Claude Code extensions" and "configuring Claude Code features". Coverage is good but a few natural phrasings (e.g., "build an agent", "write a hook", "set up permissions") are missing, so it fits anchor 4 rather than the comprehensive synonym coverage of anchor 5.

4 / 5

Distinctiveness Conflict Risk

The "Claude Code authoring kit" framing carves a clear niche with distinct triggers (specific CC feature names), so it is mostly distinct. However, "configuring Claude Code features" is broad enough to overlap with other Claude Code configuration/reference skills, keeping it at anchor 4 (minor overlap risk) rather than anchor 5.

4 / 5

Total

17

/

20

Passed

Validation

87%

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

Validation — 14 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

allowed_tools_field

'allowed-tools' contains unusual tool name(s)

Warning

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

14

/

16

Passed

Repository

modu-ai/moai-adk

Path

.claude/skills/moai-foundation-cc/SKILL.md

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.