implement-task

Implement a task with automated LLM-as-Judge verification per step

Quality

58%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./plugins/sdd/skills/implement-task/SKILL.md

Quality

Content

62%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 highly actionable and has excellent workflow clarity with explicit validation and feedback loops, but it is noticeably verbose with redundant diagrams and long example transcripts, and it makes no use of progressive disclosure — all content is inlined in a single monolithic file.

Suggestions

Remove one of the two duplicate ASCII workflow diagrams (Complete Workflow Overview vs Execution Flow Diagram) and condense the 7 narrative examples into 2-3 representative ones, moving the rest to a reference file.

Introduce progressive disclosure: split the worked examples, panel-voting algorithm, and detailed argument-resolution logic into reference files under references/ and link to them from the main body.

Trim the repeated CRITICAL/MANDATORY emphasis (e.g. consolidate the orchestrator-only table, Anti-Rationalization Rules, and Why-This-Matters into a single concise section) to reduce token cost without losing the operational rule.

Dimension	Reasoning	Score
Conciseness	At ~1947 lines the body is noticeably verbose: it includes two near-duplicate ASCII workflow diagrams (Complete Workflow Overview and Execution Flow Diagram of the same 5 phases), seven long narrative example transcripts (~290 lines), and repeated CRITICAL/MANDATORY emphasis on the same delegation point — several padded/redundant sections matching the score-2 anchor; it avoids score-1 only because it does not explain basic concepts Claude already knows.	2 / 5
Actionability	Concrete executable bash commands (git mv, ls .specs/tasks/..., git diff --cached --name-only), full agent-dispatch prompt templates, threshold-application pseudocode, and a structured retry-feedback template provide mostly executable guidance; the minor gap is that prompts rely on template placeholders ([Step N], $TASK_PATH) requiring substitution, keeping it just below fully copy-paste-ready score-5.	4 / 5
Workflow Clarity	The process is explicitly sequenced (Phase 0→5 with numbered steps 0.1-0.5, 1.1-1.3, Patterns A/B/C), with explicit validation checkpoints (per-step code-reviewer gate, orchestrator-level PASS rule), feedback loops (FAIL → retry with reviewer issues → re-verify until PASS or MAX_ITERATIONS), DoD verification in Phase 3, and a closing checklist — matching the score-5 anchor.	5 / 5
Progressive Disclosure	No bundle files exist (references/, scripts/, assets/ absent) and the entire 1947-line skill is one monolithic file; although it has strong section headers, content that clearly belongs in separate reference files (7 worked examples, panel-voting algorithm, detailed config resolution, error handling) is inlined, fitting the score-3 "some structure but content that should be separate is inline" anchor rather than score-2 (which requires minimal structure/no headers).	3 / 5
	Total	14 / 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 is concise and states a clear, somewhat distinctive capability, but it lacks an explicit "Use when..." trigger clause and natural trigger-term coverage, which cap completeness and trigger quality at 3.

Suggestions

Add an explicit trigger clause, e.g. "Use when implementing a multi-step task from a .specs task file that needs per-step automated quality verification."

Include natural trigger terms and synonyms users would actually say (e.g. "task file", ".specs", "code review", "iterate until quality threshold") to improve trigger-term coverage.

List 1-2 more concrete actions (e.g. "dispatch developer + code-reviewer sub-agents, aggregate panel scores, iterate fixes") to raise specificity above 3.

Dimension	Reasoning	Score
Specificity	Quotes "Implement a task" (generic action) plus "automated LLM-as-Judge verification per step" (one concrete mechanism) — names the domain and 1-2 concrete actions but is not comprehensive, matching the score-3 anchor rather than the multi-action score-4/5 anchors.	3 / 5
Completeness	It states a clear "what" ("Implement a task with automated LLM-as-Judge verification per step") but has no "Use when..." or equivalent trigger clause, so per the rubric guideline a missing trigger clause caps completeness at 3.	3 / 5
Trigger Term Quality	Relevant keywords ("implement", "task", "verification", "LLM-as-Judge") exist but common variations/synonyms and natural user phrasings are missing, fitting the score-3 anchor "some relevant keywords but missing common variations" rather than the broad-coverage score-4.	3 / 5
Distinctiveness Conflict Risk	The "automated LLM-as-Judge verification per step" qualifier gives it a clear niche with only minor overlap risk against generic implementation skills, fitting the score-4 "mostly distinct" anchor better than the score-3 "could still overlap" anchor.	4 / 5
	Total	13 / 20 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (1948 lines); consider splitting into references/ and linking	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	14 / 16 Passed

Repository: NeoLabHQ/context-engineering-kit
Path: plugins/sdd/skills/implement-task/SKILL.md
Commit: 8539779

Reviewed: 5 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.