workflow-lite-plan
Explore-first wave pipeline. Decomposes requirement into exploration angles, runs wave exploration via spawn_agents_on_csv, synthesizes findings into execution tasks with cross-phase context linking (E*→T*), then wave-executes via spawn_agents_on_csv.
46
36%
Does it follow best practices?
Impact
Pending
No eval scenarios have been run
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./.codex/skills/workflow-lite-planex/SKILL.mdQuality
Discovery
17%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 heavily laden with internal/technical jargon that would be opaque to most users and to Claude when selecting among skills. It lacks a 'Use when...' clause entirely, and the concrete actions are described in implementation-specific terms rather than user-facing language. The description needs significant reworking to be useful for skill selection.
Suggestions
Add an explicit 'Use when...' clause describing the user scenarios that should trigger this skill, e.g., 'Use when the user wants to break down a complex requirement into research phases followed by execution phases.'
Replace internal jargon like 'spawn_agents_on_csv', 'cross-phase context linking (E*→T*)' with plain-language descriptions of what these do, e.g., 'delegates parallel research tasks to sub-agents, then synthesizes findings into actionable execution steps.'
Include natural trigger terms a user might say, such as 'explore then execute', 'research before implementing', 'multi-phase task breakdown', or 'parallel exploration'.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description names specific actions like 'decomposes requirement into exploration angles', 'runs wave exploration', 'synthesizes findings into execution tasks', and 'wave-executes', but these are described in domain-specific jargon (e.g., 'cross-phase context linking (E*→T*)') rather than universally understood concrete actions. | 2 / 3 |
Completeness | While the description attempts to explain 'what' (explore-first wave pipeline with decomposition and execution), there is no 'Use when...' clause or any explicit guidance on when Claude should select this skill. The 'what' itself is also unclear due to jargon, making both halves weak. | 1 / 3 |
Trigger Term Quality | The description uses highly technical internal jargon like 'spawn_agents_on_csv', 'cross-phase context linking (E*→T*)', and 'wave pipeline' — none of which are natural terms a user would say when requesting this kind of work. There are no user-facing trigger terms. | 1 / 3 |
Distinctiveness Conflict Risk | The highly specific internal terminology like 'spawn_agents_on_csv' and 'E*→T*' makes it unlikely to conflict with other skills, but the overall vagueness of the domain ('requirement decomposition', 'exploration', 'execution') could overlap with general planning or task management skills. | 2 / 3 |
Total | 6 / 12 Passed |
Implementation
55%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill is extremely thorough and actionable with excellent workflow clarity, mandatory confirmation gates, and complete executable code for every phase. However, it is severely over-long for a SKILL.md file—the ~600+ lines of inline implementation detail (helper functions, full instruction templates, wave computation algorithms, report generation) should be split into referenced bundle files. The monolithic structure makes it a poor fit for token-efficient context loading.
Suggestions
Extract instruction templates (buildExploreInstruction, buildExecuteInstruction) into separate files like EXPLORE_TEMPLATE.md and EXECUTE_TEMPLATE.md, referenced from the main skill
Move utility functions (computeWaves, buildPrevContext, updateMasterCsvRow) into a referenced helper script file rather than inlining them
Move the detailed CSV schema tables and example rows into a separate SCHEMAS.md file, keeping only a brief summary in the main skill
Condense the Phase 5 context.md generation code to a brief description—Claude can generate a markdown report from a specification without seeing the full template string
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Extremely verbose at ~600+ lines. Contains massive amounts of implementation detail that could be in separate files. The full JavaScript implementations of every helper function, CSV update logic, wave computation, instruction templates, and context.md generation are all inline. Much of this (like CSV escaping, Kahn's algorithm, session initialization) is boilerplate Claude can derive. The ASCII diagrams, while helpful, add to the bulk. | 1 / 3 |
Actionability | Highly actionable with complete, executable JavaScript code for every phase. CSV schemas are fully specified with example rows, spawn_agents_on_csv calls include all parameters and output schemas, instruction templates are copy-paste ready, and helper functions are complete implementations. Every step has concrete code. | 3 / 3 |
Workflow Clarity | Excellent workflow clarity with 5 clearly sequenced phases, explicit mandatory confirmation gates between phases, dependency cascade handling (skip on failure), wave ordering rules, and validation checkpoints. The error handling table covers edge cases, and core rules explicitly state ordering constraints. Feedback loops are present (retry failed tasks in Phase 5). | 3 / 3 |
Progressive Disclosure | Monolithic wall of text with no bundle files to offload content to. The entire implementation—helper functions, instruction templates, wave computation algorithm, CSV schemas, context.md generation—is all inline in a single massive file. The instruction templates, CSV schemas, and utility functions should be in separate referenced files. | 1 / 3 |
Total | 8 / 12 Passed |
Validation
72%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 8 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
skill_md_line_count | SKILL.md is long (1145 lines); consider splitting into references/ and linking | Warning |
allowed_tools_field | 'allowed-tools' contains unusual tool name(s) | Warning |
frontmatter_unknown_keys | Unknown frontmatter key(s) found; consider removing or moving to metadata | Warning |
Total | 8 / 11 Passed | |
- Repository
- catlog22/Claude-Code-Workflow
- Commit
227244f
- Reviewed
Table of Contents
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.