CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

axiom-build

Use when ANY iOS or macOS build fails, a crash log needs diagnosing, a test run crashes, Xcode misbehaves, or an environment issue blocks work before code is the suspect. Covers build failures, dependency conflicts, simulator diagnostics.

Quality

72%

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-plugin/plugins/axiom/skills/axiom-build/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

60%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 genuinely actionable router: concrete commands, an environment-first diagnostics procedure with explicit validation checkpoints, and clearly signaled one-level-deep routing to sub-skills. Its main defect is heavy redundancy — the same routing table appears three times (Routing Logic, Decision Tree, Example Invocations) — which costs significant tokens without adding guidance, and small structural flaws (misordered section numbers 16/15) further suggest it needs tightening.

Suggestions

Collapse the redundancy: keep the Routing Logic subsections as the canonical map and replace the Decision Tree and the ~40-entry Example Invocations list with a handful of representative examples, cutting the file roughly in half.

Fix structural jumbles: renumber sections so code-signing (15) precedes xclog-ref (16) as listed, and move the dense axbuild/pgrep diagnostics procedure into a referenced file (e.g. scripts/axbuild.md) so SKILL.md stays a lean router overview.

Make the build/test code block executable by showing how to discover the scheme and destination inline (e.g. `xcodebuild -list` / `xcrun simctl list`) before the "$AXBUILD" invocation, instead of leaving $SCHEME/$DESTINATION/$PACKAGE as undefined placeholders.

DimensionReasoningScore

Conciseness

The routing map is restated three times: "Routing Logic" (16 subsections with triggers), the "Decision Tree" (16 lines repeating the same mapping), and ~115 lines of "Example Invocations", plus a "When to Use" list duplicating the description. This matches anchor 2 (several unnecessary/padded sections) — not 3 because the duplication is substantial rather than occasional, not 1 because there is no padding that explains concepts Claude already knows.

2 / 5

Actionability

Concrete, executable guidance throughout: `pgrep -lx xcodebuild; echo "pgrep exit=$?"`, the `"$AXBUILD" xcodebuild -scheme "$SCHEME" -destination "$DESTINATION" build` block, slash commands (`/axiom:fix-build`, `/axiom:console`), and explicit skill file paths. Matches anchor 4 — not 5 because the code block uses undefined placeholder variables ($SCHEME, $DESTINATION, $PACKAGE) and is not copy-paste ready, though the text does instruct discovering the actual scheme and destination first.

4 / 5

Workflow Clarity

The "Capture Build and Test Diagnostics" sequence is clearly ordered with explicit checkpoints: pgrep exit-code interpretation ("exit 1 means none are running"), "Inspect `command` (native outcome) and `collection` (evidence completeness) separately", "record its exit status", and truncation-marker checks before rebuilding. Matches anchor 4 — not 5 because document-level organization (sections numbered 14→16→15, an interleaved auto-maintained block, and three parallel routing restatements) muddies the overall flow; not 3 because validation checkpoints are explicit rather than implicit.

4 / 5

Progressive Disclosure

No bundle files exist (references/, scripts/, assets/ are absent); routing pointers such as "Invoke: `skills/xcode-debugging.md`" and "See axiom-shipping (skills/testflight-triage.md)" are one level deep and clearly signaled, an appropriate split for a router. Matches anchor 4 — not 5 because the ~115-line Example Invocations list and the dense diagnostics procedure inflate the main file where a separate reference would serve, and section ordering is jumbled; not 3 because references are clearly signaled and content is mostly well placed.

4 / 5

Total

14

/

20

Passed

Description

73%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, trigger-rich description with an explicit 'Use when...' clause and a clearly bounded iOS/macOS build-and-environment niche. The main weakness is the 'what' half: 'Covers...' names issue classes but never states what the skill actually does (environment-first diagnosis and routing to specialized sub-skills).

DimensionReasoningScore

Specificity

"Covers build failures, dependency conflicts, simulator diagnostics" names the iOS/macOS build domain and several specific issue classes, but lists few concrete actions the skill performs (no diagnosing/routing/capture verbs). Matches anchor 3: domain plus 1-2 concrete actions, not comprehensive — not 2 because multiple specific issue classes are named, not 4 because it is a coverage list rather than a list of actions.

3 / 5

Completeness

"Use when ANY iOS or macOS build fails, a crash log needs diagnosing..." gives a fully explicit 'when' with concrete triggers, but the 'what' is limited to the vague "Covers build failures, dependency conflicts, simulator diagnostics" without stating what the skill does. Both halves are present with a weaker 'what', matching anchor 4 — not 5 because the capability statement is a coverage list, not 3 because 'when' is explicit and 'what' is present.

4 / 5

Trigger Term Quality

"build fails", "crash log", "test run crashes", "Xcode misbehaves", and "simulator" are natural phrases a user would say. Good keyword coverage with a few natural terms missing ("won't compile", "linker error", "BUILD FAILED", ".ips"), matching anchor 4 rather than 5.

4 / 5

Distinctiveness Conflict Risk

"iOS or macOS build", "Xcode", "simulator", and the boundary phrase "before code is the suspect" establish a clear Apple build/environment niche with distinct triggers and minimal overlap risk, matching anchor 5; anchor 4's minor-overlap caveat fits less well since the triggers are platform-specific.

5 / 5

Total

16

/

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.

Validation — 15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

referenced_paths_exist

Referenced path issues: 1 missing

Warning

Total

15

/

16

Passed

Repository

CharlesWiltgen/Axiom

Path

.claude-plugin/plugins/axiom/skills/axiom-build/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.