CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

flutter-ci-cd

Create, repair, or harden provider-neutral Flutter CI/CD pipelines, quality gates, caches, artifacts, secrets, and staged delivery. Use for GitHub Actions, GitLab CI, Codemagic, Bitrise, Xcode Cloud, or similar orchestration; route local artifact configuration to flutter-build-release and never publish without explicit authorization.

Quality

89%

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

The canonical home for this skill is thiennc-tesoglobal/flutter-skills

SKILL.md
Quality
Evals
Security

Quality

Content

86%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.

The body is an exemplary lean orchestration document: it assumes Claude's competence, preserves repository invariants, routes detail to three real one-level-deep references, and includes a substantive verification section with explicit safety gates around publication and secrets. The only room for improvement is naming concrete linter/dry-run commands and making the step sequence explicit. Both scores of 4 reflect minor gaps, not weaknesses.

Suggestions

Name the concrete validation tools per provider in the Verification section (e.g., actionlint for GitHub Actions, glab ci lint for GitLab) so the syntax-validation step is executable without consulting a reference.

Make the operating sequence explicit (numbered preflight → route → implement → verify → report) so the workflow order does not have to be inferred from section headings.

DimensionReasoningScore

Conciseness

Every sentence is directive with zero padding: no explanation of what CI/CD or Flutter is, no library introductions — e.g., "A cache hit and miss must produce the same correctness result. Formatting and golden gates must not rewrite source or expectations." Each token carries an instruction the model could not infer on its own, matching the lean anchor-5 exemplar rather than anchor 4 (which allows minor over-explanation).

5 / 5

Actionability

For an instruction-only routing skill the guidance is concrete: exact files to inspect ("pubspec.yaml, lockfiles, SDK/version-manager files..."), explicit boundary conditions to verify ("cold and cached paths, pull-request and protected-branch conditions, cancellation/concurrency behavior"), and routing rules. It stops short of anchor 5 because no concrete commands appear in the body — e.g., "Validate provider syntax with the provider's current linter/dry-run facility" never names a tool (actionlint, glab ci lint), leaving the executable specifics entirely to the references.

4 / 5

Workflow Clarity

The section order encodes a clear sequence (Preflight → Route → Boundaries → Verification) and the Verification section supplies explicit checkpoints including a dry-run gate before external mutation ("prefer a dry run or non-production target before any authorized external mutation") and a failure-propagation rule ("must fail on a resulting diff"). It falls short of anchor 5 because the steps are implied by headings rather than explicitly sequenced, and there is no explicit validate → fix → re-run error-recovery loop.

4 / 5

Progressive Disclosure

The body is a lean overview with three well-signaled, one-level-deep references, each routed by task ("For deterministic format, analysis, ... read [quality gates]"), an explicit instruction to "Load only the references needed by the current pipeline", and all three referenced files verified to exist in the bundle with no nested internal references. This matches the anchor-5 structure of concise overview plus clearly labeled per-topic reference links.

5 / 5

Total

18

/

20

Passed

Description

88%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.

The description is a strong exemplar: concrete verb-led capability list, an explicit "Use for" trigger clause naming all major CI providers, and explicit disambiguation against sibling skills with a safety boundary ("never publish without explicit authorization"). Third-person/imperative voice is consistent and there is no fluff. The only gap is a handful of natural user phrasings (deploy/release/workflow) not covered by synonyms.

DimensionReasoningScore

Specificity

"Create, repair, or harden" names three concrete verbs applied to an enumerated list of pipeline concerns ("pipelines, quality gates, caches, artifacts, secrets, and staged delivery"), giving comprehensive coverage of the domain. It exceeds anchor 4 because no significant capability area of Flutter CI/CD work is left unlisted.

5 / 5

Completeness

Both questions are explicitly answered: the "what" ("Create, repair, or harden provider-neutral Flutter CI/CD pipelines, quality gates, caches, artifacts, secrets, and staged delivery") and the "when" ("Use for GitHub Actions, GitLab CI, Codemagic, Bitrise, Xcode Cloud, or similar orchestration") with concrete trigger phrases. It matches the anchor-5 exemplar structure exactly and is not merely implied like anchor 4.

5 / 5

Trigger Term Quality

Strong natural keywords users would actually say — "GitHub Actions, GitLab CI, Codemagic, Bitrise, Xcode Cloud", "CI/CD pipelines", "quality gates" — but a few common phrasings like "deploy", "release", or "workflow" are absent (covered only indirectly via "staged delivery" and "publish"). This sits between anchor 4 (a few natural terms missing) and anchor 5 (comprehensive synonyms), closer to 4.

4 / 5

Distinctiveness Conflict Risk

The niche is clear ("provider-neutral Flutter CI/CD") with explicit routing away from sibling skills ("route local artifact configuration to flutter-build-release"), minimizing overlap. Minor residual risk: the "Use for" clause lists bare provider names, so a non-Flutter GitHub Actions request could plausibly trigger it — mostly distinct with minor overlap, matching anchor 4 rather than anchor 5.

4 / 5

Total

18

/

20

Passed

Validation

100%

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

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository

thienqo/flutter-skills

Path

skills/flutter-ci-cd/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.