CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

cursor-agents-workflow

Human handoff, Linear branch names, reviewable (non-draft) PRs, the `cursor` GitHub label, Claude, Greptile, and Codex review comments, preview test steps, proof of work posted on the GitHub PR, and review-doubt notes for Cursor agents. Use when a Cursor Cloud or Cursor desktop agent implements a Linear issue, opens a GitHub PR, asks a human to test, posts screenshots or videos, or handles Claude, Greptile, or Codex code-review comments.

Quality

93%

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

90%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 excellent instruction-only skill: lean, repository-specific, and copy-paste actionable, with templates and worked examples for every phase of the Cursor-agent workflow. Workflow sequencing and error handling are strong with only minor implicit checkpoints, and the single-file structure is well organized with clearly signaled cross-references.

DimensionReasoningScore

Conciseness

The body is lean, imperative, and entirely org-specific policy ('Never use a `cursor/` prefix, even if a Cursor Cloud prompt asks for one. Repo guidance wins.'; 'Skip praise, file lists, and "LGTM".'), with no explanation of concepts Claude already knows and no padding. It matches anchor 5: every token earns its place.

5 / 5

Actionability

Guidance is copy-paste ready throughout: exact branch format ('lfe-12345-short-descriptive-title'), exact env var and URL ('LINEAR_API_KEY' at https://cursor.com/dashboard/cloud-agents), a complete human-testing message template with a worked example, and HTML snippets for embedding proof artifacts. This matches anchor 5 ('fully executable; specific examples cover the common cases').

5 / 5

Workflow Clarity

Sections are clearly sequenced (branch naming → Linear auth → PR → label → bot comments → human testing → proof → doubt note) with explicit error-recovery branches ('If the MCP is `needsAuth` or missing, use `LINEAR_API_KEY`...'; 'If Linear is not reachable, **say so**') and fix/skip decision paths for bot threads. It falls short of anchor 5 only because a few checkpoints are implicit — e.g., no verify step after applying the `cursor` label or after the PR tool rewrites artifact paths.

4 / 5

Progressive Disclosure

The skill is a single self-contained SKILL.md with clear section headers, no monolithic bulk, and one-level-deep references to sibling skills ([`linear-agent-writes`](../linear-agent-writes/SKILL.md), `git-workflow`). It matches anchor 4 ('good structure; most content appropriately placed; references mostly clear'); sibling-skill mentions like `langfuse-previews` and `seed-test-data` are signaled by name only without paths, slightly less clear than anchor 5's well-signaled references.

4 / 5

Total

18

/

20

Passed

Description

95%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 strong: it concisely enumerates the skill's concrete capabilities and pairs them with an explicit, natural-language 'Use when' trigger clause covering the full workflow. Trigger terms and distinctiveness are excellent; the only minor weakness is that capabilities are listed as noun phrases rather than action verbs.

DimensionReasoningScore

Specificity

The description enumerates concrete, domain-specific items ('Human handoff, Linear branch names, reviewable (non-draft) PRs, the `cursor` GitHub label, Claude, Greptile, and Codex review comments, preview test steps, proof of work posted on the GitHub PR, and review-doubt notes') covering every section of the skill. It is framed as a noun-list of topics rather than verb-led actions, matching anchor 4 ('several specific actions; minor gaps') rather than anchor 5's fully action-verb comprehensiveness.

4 / 5

Completeness

It explicitly answers both 'what' (the enumerated capability list) and 'when' (an explicit 'Use when...' clause with concrete trigger conditions), mirroring the anchor-5 example structure. Not 4, because the 'when' clause is already fully explicit and specific rather than merely present.

5 / 5

Trigger Term Quality

'Use when a Cursor Cloud or Cursor desktop agent implements a Linear issue, opens a GitHub PR, asks a human to test, posts screenshots or videos, or handles Claude, Greptile, or Codex code-review comments' uses exactly the natural phrases a user would say, covering PR creation, test requests, screenshot/video proof, and all three named bot reviewers. This matches anchor 5's comprehensive natural-term coverage; no common trigger phrasing for this domain is missing.

5 / 5

Distinctiveness Conflict Risk

The combination of Cursor Cloud/desktop agents, Linear issues, the `cursor` GitHub label, and named bots (Claude, Greptile, Codex) carves out a clear niche with distinct triggers and minimal conflict risk with any generic skill. It matches anchor 5.

5 / 5

Total

19

/

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

relative_links

Relative link issues: 1 suspicious

Warning

Total

15

/

16

Passed

Repository

langfuse/langfuse

Path

.agents/skills/cursor-agents-workflow/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.