CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

pr

Create a draft pull request for the Vision-Agents repo using gh CLI.

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 ./.claude/skills/pr/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

93%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, imperative instruction set: exact commands, explicit conditions, clear before/creating sequencing, and a stop-and-ask checkpoint. The only notable gap is the absence of failure handling for the pre-flight check — what to do when `dev.py check` fails is left implicit.

Suggestions

Add one line for the failure path, e.g. "If `dev.py check` fails, fix the issues and re-run it before creating the PR" — this would close the feedback-loop gap and lift workflow_clarity from 4 to 5.

DimensionReasoningScore

Conciseness

The body is pure imperative instruction — "Run `git log main..HEAD --oneline`", "Always `gh pr create --draft`. Push the branch first." — with zero padding and no explanations of concepts Claude already knows. The one rationale included ("GitHub renders each newline inside a paragraph as a visible line break") is non-obvious project convention and earns its tokens, matching the lean-and-efficient anchor at 5.

5 / 5

Actionability

Every instruction is copy-paste ready: exact commands (`git log main..HEAD --oneline`, `uv run --no-sync dev.py check`, `gh pr create --draft`), exact file paths (`.github/pull_request_template.md`, `CHANGELOG.md`), and explicit skip conditions ("Skip if the diff does not touch Python code (`*.py`) or `pyproject.toml`"). This matches the fully-executable anchor at 5 rather than 4, where only minor gaps would remain.

5 / 5

Workflow Clarity

A clear two-phase sequence ("## Before creating" → "## Creating") with a validation step (`dev.py check`) and an explicit STOP-and-ask checkpoint for multiple logical changes, matching anchor 4. It is not 5 because there is no feedback loop if `dev.py check` fails — no instruction to fix issues and re-run before creating the PR; PR creation is not destructive, so the cap-at-3 rule does not apply.

4 / 5

Progressive Disclosure

The skill is under 50 lines with no need for external bundle files (no references/, scripts/, or assets/ exist), and it is organized into two well-signaled sections; the files it does mention (`.github/pull_request_template.md`, `CLAUDE.md`, `CHANGELOG.md`) are real repo files the skill must consult, not buried references. Per the simple-skill guideline this scores 5.

5 / 5

Total

19

/

20

Passed

Description

53%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 states a single clear, concrete capability with the right tool, but it is one action only and lacks any trigger guidance, so a user or Claude has no explicit 'when to use this' signal. It is distinguishable mainly through repo scoping rather than through distinctive trigger terms.

Suggestions

Add an explicit trigger clause, e.g. "Use when asked to open/create/submit a PR (or draft PR) in the Vision-Agents repo" — this would lift completeness from 3 toward 4-5.

Include natural synonyms users actually say, such as "PR", "open a PR", and "submit a pull request", to improve trigger-term coverage from 3 to 4.

If the skill also covers pre-PR duties implied by the body (checking for multiple logical changes, running dev.py check, updating CHANGELOG.md), name them in the description to raise specificity.

DimensionReasoningScore

Specificity

"Create a draft pull request for the Vision-Agents repo using gh CLI" names the domain and one concrete action with its tool, but does not list multiple capabilities, matching 'Names domain and 1-2 concrete actions, but not comprehensive' rather than the several-actions anchor at 4.

3 / 5

Completeness

The 'what' is clear ("Create a draft pull request ... using gh CLI") but there is no 'Use when...' clause or equivalent trigger guidance, which caps completeness at 3 per the judging guidelines. It is not 4 because the 'when' is entirely absent rather than merely imprecise.

3 / 5

Trigger Term Quality

"pull request" and "draft" are natural trigger phrases, but common variations users actually say — "PR", "open a PR", "submit a PR" — are absent, matching 'Some relevant keywords but missing common variations or synonyms' rather than the good-coverage anchor at 4.

3 / 5

Distinctiveness Conflict Risk

Scoping to "the Vision-Agents repo" and to draft PRs via "gh CLI" gives it a mostly distinct niche with only minor overlap risk against generic PR/review skills, matching anchor 4. It falls short of 5 because the trigger phrasing alone would not fully distinguish it from a general PR-creation skill.

4 / 5

Total

13

/

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

GetStream/Vision-Agents

Path

.claude/skills/pr/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.