subagent-driven-development

Use when executing implementation plans with independent tasks in the current session

Quality

38%

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/subagent-driven-development/SKILL.md

Quality

Content

48%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 skill describes a well-structured subagent-driven development workflow with clear sequencing and review gates, but suffers from significant verbosity (especially the Graphviz diagrams, lengthy example, and advantages/cost sections that don't add actionable value). The core actionability is undermined by missing bundle files for the three prompt templates that are central to the workflow. The content would benefit greatly from trimming persuasive/explanatory content and ensuring the referenced prompt templates are available.

Suggestions

Remove or drastically condense the 'Advantages' and 'Cost' sections — Claude doesn't need to be persuaded about the methodology's benefits, it needs to know how to execute it.

Provide the three referenced prompt template files (implementer-prompt.md, spec-reviewer-prompt.md, code-quality-reviewer-prompt.md) as bundle files, or inline their key content, since they are essential for actionability.

Replace the verbose Graphviz dot diagrams with concise numbered step lists — the process diagram duplicates the textual description and consumes significant tokens.

Condense the example workflow to show just one task cycle (not two) and trim the conversational format to focus on the key decision points and review outcomes.

Dimension	Reasoning	Score
Conciseness	The skill is significantly verbose. The Graphviz dot diagrams are bulky and redundant with the textual descriptions. The 'Advantages' section extensively explains benefits Claude doesn't need persuading about. The 'Example Workflow' section is very long and could be condensed. The 'Red Flags' section has many items that overlap or could be consolidated. Much of this content restates the same core idea (fresh subagent + two-stage review) repeatedly.	2 / 5
Actionability	The skill references prompt templates (./implementer-prompt.md, ./spec-reviewer-prompt.md, ./code-quality-reviewer-prompt.md) but these are not provided in the bundle, making the core mechanism non-executable. The process is described at a conceptual level with an example walkthrough, but lacks concrete commands or copy-paste ready dispatch instructions. The TodoWrite usage is mentioned but not shown concretely.	3 / 5
Workflow Clarity	The multi-step process is clearly sequenced with explicit validation checkpoints (spec compliance review → code quality review → re-review loops). The ordering constraint (spec before quality) is explicitly called out. The feedback loops for reviewer issues are well-defined. Minor gap: no explicit guidance on what to do if the final cross-task review finds integration issues between tasks.	4 / 5
Progressive Disclosure	The skill references three prompt template files (./implementer-prompt.md, ./spec-reviewer-prompt.md, ./code-quality-reviewer-prompt.md) and several other skills, which is good structure. However, no bundle files are provided, so the core mechanism is incomplete. The SKILL.md itself inlines a lot of content (advantages, red flags, long example) that could be trimmed or separated, while the essential prompt templates that would make this actionable are missing.	3 / 5
	Total	12 / 20 Passed

Description

28%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.

This description is a 'when' clause without a 'what' clause. It tells Claude when to select the skill but fails to describe what the skill actually does or what concrete actions it performs. The language is abstract and uses internal jargon rather than natural user-facing trigger terms.

Suggestions

Add a clear 'what' statement describing concrete actions (e.g., 'Executes multiple independent coding tasks in parallel by spawning sub-agents for each task').

Include natural trigger terms users might say, such as 'parallel tasks', 'batch implementation', 'run tasks simultaneously', or 'implement multiple changes'.

Specify the scope more clearly—what kinds of tasks? Code changes, file operations, test runs? This would reduce ambiguity and conflict risk.

Dimension	Reasoning	Score
Specificity	Names a domain ('executing implementation plans with independent tasks') but provides no concrete actions. We don't know what 'executing' entails—writing code, running commands, creating files, etc.	2 / 5
Completeness	Has a 'when' clause ('Use when executing implementation plans...') but the 'what' is essentially absent—there's no description of what the skill actually does, only when to use it.	2 / 5
Trigger Term Quality	The phrase 'implementation plans with independent tasks' is somewhat technical/internal jargon. Users are unlikely to naturally say 'execute implementation plans with independent tasks'—they might say 'run tasks in parallel', 'implement the plan', or 'do the tasks'.	2 / 5
Distinctiveness Conflict Risk	The phrase 'independent tasks in the current session' provides some specificity that distinguishes it from general task execution, but 'implementation plans' is broad enough to overlap with many coding/planning skills.	3 / 5
	Total	9 / 20 Passed

Validation

90%

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 — 10 / 11 Passed

Validation for skill structure

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

	Total	10 / 11 Passed

Repository: administrakt0r/AI-Agents-Safe-Coding-Skills
Path: skills/subagent-driven-development/SKILL.md
Commit: 57c135b

Reviewed: 15 hours 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.