CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

setup-osn

Use when setting up or repairing a development environment for this repository on a machine — Bun at the pinned version, the GitHub CLI and its auth, SSH commit signing and the recommended git config, workspace dependencies, lefthook hooks, the TypeScript language server agents use and the portless dev proxy — and verifying it with the type-check, lint and format gates. Also the place to start when bun, gh, turbo or a hook is missing or the wrong version.

76

Quality

96%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

SKILL.md
Quality
Evals
Security

Quality

Content

92%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.

An exemplary operational skill body: executable commands in every step, a contingency matrix up front, validation gates and honesty rules throughout, and deliberate version-deferral to the repository so nothing goes stale. The only blemish is the time-sensitive 'as of this writing' hook listing in Step 7, which duplicates what the skill itself says to read from lefthook.yml.

Suggestions

Trim the 'as of this writing pre-commit runs oxlint and `bun fmt`… pre-push runs the type check, `bun audit`…' sentence in Step 7 to just the instruction to read lefthook.yml and tell the user; the inline list duplicates the file and will go stale at the next hook change.

The long bootstrap-bare-root paragraph in Step 7 could be reduced to what to do and what to report, moving the rationale about worktree-subdirectory sessions into the script's own output or a wiki page it already links.

Tighten a few rationale clauses (e.g., 'which is why portless is tilde-pinned in package.json — a version bump is a change to review') to the actionable core where the reason is already obvious to an experienced agent.

DimensionReasoningScore

Conciseness

Efficient and assumes Claude's competence — no explanations of Homebrew, SSH, or GPG basics, and every pinned version is deferred to the tree ("Every pinned value comes from the repository") — but Step 7 inline-lists time-sensitive hook contents ("as of this writing pre-commit runs oxlint and `bun fmt` in write mode with `stage_fixed`, and pre-push runs the type check, `bun audit --audit-level=high`…") immediately after saying "Read `lefthook.yml` for what the hooks do", which is duplicated, staled-at-bump detail that could be trimmed. This sits between the 4 and 5 anchors, closer to efficient-with-minor-trimmable-instances.

4 / 5

Actionability

Fully executable, copy-paste-ready commands throughout — `curl -fsSL https://bun.sh/install | bash -s "bun-v$(cat .bun-version)"`, the `npm install -g --before=…` soak line, a check/apply git-config table, `bash scripts/setup.sh`, `bunx portless proxy start` — with the common cases (macOS, non-macOS, no sudo, no network, non-interactive) explicitly covered, matching the score-5 anchor.

5 / 5

Workflow Clarity

Nine clearly ordered steps with explicit validation checkpoints (`gh auth status`, `gh repo view`, `gh pr list`, re-verify after install, the three gates, `bunx portless doctor`, `claude plugin list`), feedback loops ("If it does not print the version… install that exact version. Verify again"; "If a gate fails, paste the shortest decisive lines into the table and keep going"), and a hard honesty rule ("Never claim a gate passed that did not run here") plus a mandated final report table — the score-5 anchor.

5 / 5

Progressive Disclosure

No bundle files exist, and the body stays lean by deferring detail to clearly-signaled, one-level-deep repository files (`lefthook.yml`, `scripts/setup.sh`, `wiki/conventions/devloop-urls.md`, `wiki/conventions/agent-tooling.md`) under the rule "If any of those disagree with what is below, the tree wins" — well-organized step headers and a contingency section make navigation easy, matching the score-5 anchor.

5 / 5

Total

19

/

20

Passed

Description

100%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 model description: concrete, comprehensive enumeration of capabilities, explicit and varied when-triggers including the failure modes a user would actually voice, third-person-appropriate imperative style, and a clear repository-specific niche. Density is high but every clause carries trigger or capability information.

DimensionReasoningScore

Specificity

Enumerates every concrete capability — "Bun at the pinned version, the GitHub CLI and its auth, SSH commit signing and the recommended git config, workspace dependencies, lefthook hooks, the TypeScript language server agents use and the portless dev proxy — and verifying it with the type-check, lint and format gates" — matching the score-5 anchor for multiple specific concrete actions with comprehensive coverage; there is no gap that would drop it to 4.

5 / 5

Completeness

Explicitly answers both what (enumerate-and-configure the listed components, then verify with the three gates) and when ("Use when setting up or repairing a development environment… Also the place to start when bun, gh, turbo or a hook is missing or the wrong version"), matching the score-5 anchor with concrete trigger phrases; the when-clauses are explicit, not the score-4 anchor's 'could be more specific'.

5 / 5

Trigger Term Quality

Natural phrases a user would say ("setting up or repairing a development environment", "missing or the wrong version") plus exact tool names as triggers ("bun, gh, turbo or a hook") give comprehensive natural-term coverage including tool-name synonyms, matching the score-5 anchor rather than the 4 anchor's 'a few natural terms missing'.

5 / 5

Distinctiveness Conflict Risk

Scoped to "this repository" and to a named toolchain niche with distinct triggers (bun/gh/turbo/lefthook/portless), so it is clearly distinguishable from other skills with minimal conflict risk — the score-5 anchor; it is not merely 'mostly distinct' as at 4.

5 / 5

Total

20

/

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: 4 missing

Warning

Total

15

/

16

Passed

Repository

englishstreetventures/osn

Path

.claude/skills/setup-osn/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.