task

Work a task end-to-end with lean context gathering, implementation, and verification

Quality

61%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./.agents/skills/task/SKILL.md

Quality

Content

77%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.

This is a highly actionable, well-sequenced workflow skill with strong validation gates and copy-paste-ready commands throughout, though the ~500-line body is monolithic and repeats several core rules across up to five sections. Splitting domain-specific procedures into reference files and deduplicating the PR-plan rules would fix its main weaknesses.

Suggestions

Move the Security Advisory Hotfixes procedure and the Task-Style PR Body template into separate reference files (e.g. references/security-advisory.md, references/pr-body.md) linked one level deep, and load them only when those sources apply — this would cut the main body substantially.

Deduplicate the PR-plan rule (state it once in Verification or the PR Body section instead of restating it in Core Rules, Intake, Verification, PR Body, and Success Criteria) and merge the `--with <pack>` catalog that appears in both Intake step 9 and the Skill Diet `autogoal` bullet.

Trim the Success Criteria section, which largely restates gates already defined in Intake, Review, and Verification, down to the few outcomes not already checkable elsewhere.

Dimension	Reasoning	Score
Conciseness	The body is dense and repo-specific with essentially no explanation of concepts Claude already knows, but the same rules are restated repeatedly — the PR-plan/`🧭 Task plan` requirement appears in Core Rules, Intake, Verification, Task-Style PR Body, and Success Criteria, and the `--with <pack>` catalog appears in both Intake step 9 and the Skill Diet `autogoal` bullet. It is mostly efficient but could be meaningfully tightened, fitting the anchor exactly; it is not 2 because there is no concept-explaining filler, and not 4 because the duplication across sections is more than minor.	3 / 5
Actionability	The guidance is fully executable with copy-paste-ready commands and exact flags: `gh issue view`, `gh api repos/<owner>/<repo>/security-advisories/<GHSA_ID>`, `node .agents/skills/autogoal/scripts/create-goal-scratchpad.mjs --template <task\|docs> --with <pack> --title "..."`, `npm view <package>@<version>`, `pnpm run reinstall`, and `gh pr view --json body`. Concrete examples (e.g. `🐛 Fixes #123`, `🟢 95-100% confidence`, the exact table header) cover the common cases; as an instruction-only skill this fully satisfies the rubric's code-vs-instruction note.	5 / 5
Workflow Clarity	The multi-step process is explicitly sequenced (14-step numbered Intake, per-shape Execution Paths, Verification, Final Handoff) with explicit validation checkpoints and feedback loops: reproduce-before-fix with a four-level escalating repro ladder, 'run `pnpm run reinstall` once and rerun the exact failing command', autoreview loop 'keep going until there are no accepted/actionable findings', and hard-stop verdicts (`not reproduced`, `invalid`) before code. Batch work is covered with per-PR validation rather than aggregate evidence, so the batch cap does not apply.	5 / 5
Progressive Disclosure	Section headers are clear and well-ordered, but the skill is a ~500-line monolith with no bundle files at all: content that clearly belongs in separate one-level-deep references — the Security Advisory Hotfixes procedure, the Task-Style PR Body template, the Skill Diet catalog — is all inlined. This fits 'some structure but content that should be separate is inline'; it is above 2 because navigation via headers is genuinely easy, and below 4 because no reference files exist to split the bulk into.	3 / 5
	Total	16 / 20 Passed

Description

46%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 clear what with terse phase-level actions, but lacks any 'Use when...' trigger guidance and is so broad that it would collide with most other coding-workflow skills. It reads as an internal routing label rather than a user-facing trigger description.

Suggestions

Add an explicit 'Use when...' clause, e.g. 'Use when the user gives a task description, bug, feature request, issue/PR link, or asks for end-to-end implementation and verification.'

Include natural trigger terms and synonyms users would actually say — bug fix, feature, refactor, GitHub issue, PR, tracker item — to raise trigger-term quality and distinctiveness.

Distinguish the skill from sibling workflow skills (e.g. `major-task`) in the description itself, since the body routes between them but the description gives no signal for choosing this one.

Dimension	Reasoning	Score
Specificity	The description names the domain (task work) and three actions — 'lean context gathering, implementation, and verification' — but these are high-level phases rather than comprehensive concrete actions, matching the anchor for domain plus 1-2 concrete actions. It is not score 4 because no specific operations (e.g., bug fixes, PR creation, issue triage) are listed, and not score 2 because the phases do go beyond a bare domain label.	3 / 5
Completeness	It has a clear 'what' ('work a task end-to-end with lean context gathering, implementation, and verification') but no 'when' — there is no 'Use when...' clause or equivalent explicit trigger guidance, which caps completeness at 3 per the judging guidelines. It is not 2 because the 'what' is clear, and not 4 because the 'when' is entirely absent rather than weakly implied.	3 / 5
Trigger Term Quality	'task', 'implementation', and 'verification' are terms a user might naturally say, but common variations a user would actually use to invoke this — bug, feature, issue, PR, refactor — are absent. It fits 'some relevant keywords but missing common variations or synonyms'; it is above score 2 (not merely generic) but below 4 (no synonym coverage).	3 / 5
Distinctiveness Conflict Risk	'Work a task end-to-end' describes virtually every engineering workflow skill, creating high overlap risk with any implementation, planning, or review skill (the body itself references near-duplicates like `major-task`). It is not 1 because the phase framing gives it slight specificity beyond 'helps with code', but it clearly falls below the midpoint toward 'very broad; high overlap risk with many similar skills'.	2 / 5
	Total	11 / 20 Passed

Validation

68%

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 — 11 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (511 lines); consider splitting into references/ and linking	Warning
metadata_version	'metadata.version' is missing	Warning
metadata_field	'metadata' should map string keys to string values	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
relative_links	Relative link issues: 1 suspicious	Warning

	Total	11 / 16 Passed

Repository: udecode/plate
Path: .agents/skills/task/SKILL.md
Commit: babb3c2

Reviewed: 2 days 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.