recipe-fullstack-build

Execute decomposed fullstack tasks with layer-aware agent routing

Quality

56%

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 ./skills/recipe-fullstack-build/SKILL.md

Quality

Content

77%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is a tightly written orchestration recipe with concrete routing tables, explicit status-field branches, and strong validation/feedback loops across the task cycle and post-implementation verification. Its weaknesses are verbatim repetition of the task-exclusion patterns and a fully inline structure with no local progressive-disclosure bundle files.

Suggestions

De-duplicate the exclusion pattern list by defining it once (e.g. in a Consumed Task Set definition) and referencing it from Work Plan Resolution rather than repeating the five glob patterns verbatim.

Move the stable routing tables and the Work Plan Resolution rules into a local references file (e.g. references/monorepo-flow.md) and link to it, so SKILL.md stays an overview and progressive disclosure can reach 3.

Confirm the referenced references/monorepo-flow.md actually ships in this skill's bundle or restate the essential routing rules inline if it belongs to another skill.

Dimension	Reasoning	Score
Conciseness	The body is operational and avoids explaining concepts Claude already knows, but the Work Plan Resolution and Consumed Task Set sections repeat the same exclusion patterns verbatim ("-task-prep-.md", "_overview-*.md", etc.), which could be tightened, matching the score-2 anchor. It is not a 3 due to this repetition, and not a 1 because there is no conceptual fluff or padding.	2 / 3
Actionability	The guidance is concrete and copy-paste ready: exact subagent_type values ("dev-workflows:task-executor"), a routing table, specific status-field branches ("escalation_needed", "stub_detected", "approved"), and precise file patterns. As an instruction-only orchestration skill the absence of code is not penalized because the guidance is fully actionable, matching the score-3 anchor. It is not a 2 because no key details are missing or pseudocode-only.	3 / 3
Workflow Clarity	The 4-step task cycle has explicit validation checkpoints and feedback loops (escalation checks, quality-fixer approval gates, "Proceed to next task only after layer-appropriate quality-fixer returns approved", and a post-implementation fix cycle that repeats until all pass or blocked), matching the score-3 anchor. It is not a 2 because checkpoints are explicit rather than implicit, and not a 1 because the sequence and validation are clearly defined.	3 / 3
Progressive Disclosure	No local bundle files exist (references/scripts/assets absent), so all detail is inline in SKILL.md; the only reference is to an external skill's file (subagents-orchestration-guide's monorepo-flow.md), not this skill's bundle. Sections are well-organized but content that could be split out (routing tables, work plan resolution rules) is inline, matching the score-2 anchor. It is not a 3 because there are no well-signaled one-level-deep local references, and not a 1 because sections are organized rather than a nested-reference wall of text.	2 / 3
	Total	10 / 12 Passed

Description

35%

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 domain and approach but relies on internal jargon ("layer-aware agent routing", "decomposed fullstack tasks") rather than natural user trigger terms, and omits any explicit "Use when..." guidance. It is distinguishable but not optimized for discovery or trigger matching.

Suggestions

Add an explicit trigger clause, e.g. ". Use when the user asks to execute or run a batch of pre-decomposed fullstack task files in docs/plans/tasks/."

Replace internal jargon with natural phrasing users would say, such as "run fullstack build tasks" or "execute backend and frontend task files" instead of "layer-aware agent routing".

Enumerate concrete actions (e.g. "decompose plans, route backend/frontend tasks to executors, run quality fixers, commit, and verify") to raise specificity to a 3.

Dimension	Reasoning	Score
Specificity	The phrase "Execute decomposed fullstack tasks with layer-aware agent routing" names the fullstack domain and an approach (decomposed execution, layer-aware routing), but uses abstract jargon rather than listing multiple concrete actions, matching the score-2 anchor. It is not a 3 because no concrete sub-actions are enumerated, and not a 1 because it does name a specific domain and method rather than being vague.	2 / 3
Completeness	It states what the skill does (execute decomposed fullstack tasks with layer-aware agent routing) but provides no "Use when..." clause or equivalent explicit trigger guidance, so completeness is capped at 2 per the rubric guidelines. It is not a 3 because the "when" is entirely missing, and not a 1 because the "what" is stated and not very weak.	2 / 3
Trigger Term Quality	The terms "decomposed fullstack tasks" and "layer-aware agent routing" are internal technical jargon that a user would not naturally say when they need this skill, matching the score-1 anchor of technical jargon with no natural keywords. It is not a 2 because there are no common natural-language variations a user would actually speak.	1 / 3
Distinctiveness Conflict Risk	"Decomposed fullstack tasks" with "layer-aware agent routing" targets a specific orchestration niche, but "fullstack" is broad and the phrasing is internal jargon that could overlap with similar orchestration skills, matching the score-2 anchor. It is not a 3 because it lacks clear distinct natural triggers, and not a 1 because it is more specific than generic phrases like "Helps with code and documents."	2 / 3
	Total	7 / 12 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
referenced_paths_exist	Referenced path issues: 1 missing	Warning

	Total	15 / 16 Passed

Repository: shinpr/claude-code-workflows
Path: skills/recipe-fullstack-build/SKILL.md
Commit: 56ab6c1

Reviewed: 4 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.