start-work

Start work on a Jira ticket.

Quality

74%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

—

The risk profile of this skill

Fix and improve this skill with Tessl

tessl review fix ./.claude/skills/start-work/SKILL.md

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 instruction-only skill: correct step ordering, dense domain specifics (IDs, transition sequences, patterns), and explicit guardrails for the risky preconditions. Its main gaps are the unspecified Jira invocation mechanism and the absence of end-state verification after the transitions and branch creation.

Suggestions

Specify how to talk to Jira (e.g., the CLI/API command or MCP tool to fetch issues and apply transition IDs) so the guidance is fully executable.

Add end-state verification steps: confirm the target is In Progress and correctly assigned, and confirm the branch exists/checked out, before entering plan mode.

Clarify how long/how often to refetch the parent's subtasks waiting for the autocreated Technical Task, and what to do if it never appears.

Dimension	Reasoning	Score
Conciseness	The ~44-line body is lean imperative prose with zero padding — no explanation of what Jira or git branches are, and dense specifics like the two transition tables earn every token. It matches anchor 5 ('lean and efficient; assumes Claude's competence') with nothing identifiable to trim.	5 / 5
Actionability	Concrete, executable guidance throughout: issue-type IDs (10004, 10001, 10002, 10153), transition IDs (61, 41 then 61, 21), the key pattern [A-Z]+-[0-9]+, and the refetch-until-appears instruction. It falls short of anchor 5 because no literal Jira commands/API calls are given (the invocation mechanism is unspecified) and 'Make a Plan' is direction only; it is well above anchor 3's pseudocode level.	4 / 5
Workflow Clarity	A clear multi-step sequence with explicit guardrails ("Abort when the working tree has uncommitted changes", "refuse to continue" when the parent is in progress by another user) and a polling loop for the autocreated subtask. Score 5 is not reached because end-state checkpoints are missing — no verification that transitions applied or that the target/branch reached the desired final state.	4 / 5
Progressive Disclosure	The skill is under 50 lines, single-purpose, requires no external references, and is organized into well-ordered sections mirroring the actual workflow order. Per the rubric's simple-skill guideline, this earns anchor 5: clear organization with nothing that belongs in a separate file inlined.	5 / 5
	Total	18 / 20 Passed

Description

52%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 appropriately concise, in third person, and unambiguously identifies its Jira niche, but it undersells the skill's capabilities. It states only one generic action and omits any 'when to use' trigger guidance, so it functions as a label rather than an activatable description.

Suggestions

List the concrete actions the skill performs, e.g. "Assign a Jira ticket, transition it to In Progress, create a git branch, and enter plan mode."

Add an explicit trigger clause, e.g. "Use when the user asks to start, begin, or pick up work on a Jira ticket, or provides a ticket key like LPD-86295 or a Jira browse URL."

Include natural trigger variations and formats users actually mention: ticket keys, Jira URLs, and phrases like "start working on this ticket".

Dimension	Reasoning	Score
Specificity	"Start work on a Jira ticket." names the domain (Jira tickets) but offers only a single generic action with no concrete sub-actions (assign, transition, branch, plan), matching the 'names the domain but actions are minimal or generic' anchor. It is not entirely abstract, so score 1 does not fit.	2 / 5
Completeness	The description has a clear 'what' ("Start work on a Jira ticket") but no 'Use when...' or equivalent 'when' trigger clause, which per the judging guidelines caps completeness at 3. It matches anchor 3 exactly: clear 'what', 'when' missing.	3 / 5
Trigger Term Quality	"Jira" and "ticket" are natural, domain-specific terms a user would actually say, but common variations ("begin working on", "pick up this ticket") and URL/keyword triggers are missing. This lands on 'some relevant keywords but missing common variations or synonyms' rather than the thinner anchor-2 coverage.	3 / 5
Distinctiveness Conflict Risk	"Jira ticket" carves out a clear niche with distinct triggers and minimal conflict risk — no plausible competing skill would claim starting Jira ticket work. This matches anchor 5 ('clear niche with distinct triggers; minimal conflict risk'); anchor 4's 'minor overlap risk' is not warranted.	5 / 5
	Total	13 / 20 Passed

Validation

93%

Warnings & errors only

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

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: liferay/liferay-portal
Path: .claude/skills/start-work/SKILL.md
Commit: 994108c

Reviewed: 1 day ago

Table of Contents

Discovery Implementation Validation

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.