workflow-multi-cli-plan
Multi-CLI collaborative planning with codebase context gathering, iterative cross-verification, and execution handoff.
35
31%
Does it follow best practices?
Impact
—
No eval scenarios have been run
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./.claude/skills/workflow-multi-cli-plan/SKILL.mdQuality
Discovery
7%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 abstract jargon and buzzwords without conveying concrete actions or use cases. It lacks a 'Use when...' clause, natural trigger terms, and specific capability descriptions, making it nearly impossible for Claude to reliably select this skill from a pool of alternatives.
Suggestions
Replace abstract terms with concrete actions, e.g., 'Coordinates multiple Claude CLI instances to plan and execute complex coding tasks across a codebase' instead of 'Multi-CLI collaborative planning with codebase context gathering'.
Add an explicit 'Use when...' clause with natural trigger terms, e.g., 'Use when the user wants to coordinate multiple agents, plan a large refactor across files, or needs parallel codebase analysis and execution.'
Include natural keywords users would actually say, such as 'multi-agent', 'parallel tasks', 'large refactor', 'coordinate agents', 'plan and execute', or 'complex multi-file changes'.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description uses abstract, buzzword-heavy language like 'collaborative planning', 'codebase context gathering', 'iterative cross-verification', and 'execution handoff' without describing concrete actions a user would understand. No specific tasks are listed. | 1 / 3 |
Completeness | The 'what' is vague and abstract, and there is no 'when' clause or explicit trigger guidance at all. It fails to answer either question clearly. | 1 / 3 |
Trigger Term Quality | The terms used ('Multi-CLI', 'cross-verification', 'execution handoff') are technical jargon that users would almost never naturally say. There are no natural keywords a user would use when requesting this skill. | 1 / 3 |
Distinctiveness Conflict Risk | The 'Multi-CLI' aspect provides some distinctiveness, but terms like 'collaborative planning' and 'codebase context gathering' are broad enough to overlap with many development-related skills. It's somewhat specific to a multi-agent workflow but not clearly delineated. | 2 / 3 |
Total | 5 / 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.
The skill excels at actionability and workflow clarity with concrete, executable code for every phase and a well-structured multi-phase workflow with convergence checks and feedback loops. However, it is severely hampered by poor token efficiency — the entire content is a monolithic wall of code that inlines agent prompts, schemas, and detailed assembly logic that should be split into referenced files. The lack of progressive disclosure means this skill consumes enormous context window space on every invocation.
Suggestions
Extract the synthesis.json schema, agent prompt templates, and executionContext assembly into separate referenced files (e.g., schemas/synthesis.json, prompts/discuss-agent.md, prompts/planning-agent.md) to dramatically reduce the SKILL.md token footprint.
Move the full code examples for Phase 5 (context-package building, planning agent invocation, executionContext assembly) into a referenced EXECUTION.md file, keeping only a concise summary in the main skill.
Remove redundant information — the context-package is defined in Step 1 and then repeated verbatim inside the planning agent prompt; reference it instead.
Add a concise 'Quick Reference' section at the top summarizing the 5 phases in 5-10 lines, then link to detailed phase documentation in separate files.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is extremely verbose at ~300+ lines with extensive inline code blocks that could be referenced externally. Many sections repeat information (e.g., the context-package is built and then passed inline to the agent prompt). The synthesis.json schema, full agent prompts, and detailed executionContext assembly are all inlined rather than referenced, creating significant token bloat. | 1 / 3 |
Actionability | The skill provides fully concrete, executable JavaScript code for every phase including session initialization, agent invocation with complete prompt templates, convergence decision logic, user question formatting, plan generation, and execution handoff. Code examples are copy-paste ready with specific variable names, file paths, and JSON structures. | 3 / 3 |
Workflow Clarity | The five-phase workflow is clearly sequenced with an ASCII flow diagram, explicit validation/convergence checks between phases, feedback loops (Phase 4 → Phase 2 for more analysis), completion checklists for agents, and clear routing logic. The convergence decision tree and error handling table provide robust checkpoint coverage. | 3 / 3 |
Progressive Disclosure | Everything is crammed into a single monolithic file with no bundle files provided. The synthesis.json schema, full agent prompts, context-package assembly, and execution handoff code should be in separate referenced files. The skill references external schemas (plan-overview-base-schema.json, task-schema.json) but doesn't reference any companion documentation files for its own extensive content. | 1 / 3 |
Total | 8 / 12 Passed |
Validation
90%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 |
|---|---|---|
allowed_tools_field | 'allowed-tools' contains unusual tool name(s) | Warning |
Total | 10 / 11 Passed | |
- Repository
- catlog22/Claude-Code-Workflow
- Commit
5ff5e86
- 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.