large-workspace-handling
Rosetta skill to partition large workspaces or folders (100+ files recursively) into scoped subagent tasks when single-agent context is insufficient.
40
40%
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 ./instructions/r2/core/skills/large-workspace-handling/SKILL.mdQuality
Discovery
57%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 identifies a clear and distinctive niche (partitioning large workspaces for subagent delegation) but lacks an explicit 'Use when...' clause and relies on some jargon that users wouldn't naturally use. It would benefit from listing more concrete actions and including natural trigger terms that match how users would phrase their requests.
Suggestions
Add an explicit 'Use when...' clause, e.g., 'Use when the user needs to process a large codebase, split work across subagents, or when a folder has too many files to fit in a single context window.'
Include more natural trigger terms users would say, such as 'too many files', 'split work', 'large codebase', 'divide tasks', 'parallelize', or 'break down project'.
List additional concrete actions beyond 'partition', e.g., 'scans directory trees, groups files by module or dependency, generates scoped task descriptions for subagents'.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | It names the domain (partitioning large workspaces) and a key action (partition into scoped subagent tasks), but doesn't list multiple concrete actions—e.g., it doesn't specify how partitioning works, what outputs are produced, or what strategies are used. | 2 / 3 |
Completeness | It answers 'what' (partition large workspaces into subagent tasks) reasonably well, but there is no explicit 'Use when...' clause or equivalent trigger guidance. The 'when' is only implied by the condition '100+ files recursively' and 'single-agent context is insufficient'. | 2 / 3 |
Trigger Term Quality | Includes some relevant terms like 'large workspaces', '100+ files', 'subagent tasks', and 'context', but uses jargon like 'Rosetta skill' and 'single-agent context is insufficient' that users wouldn't naturally say. Missing common trigger phrases like 'split work', 'too many files', 'divide and conquer', or 'parallelize'. | 2 / 3 |
Distinctiveness Conflict Risk | The description carves out a clear niche—partitioning large file sets into subagent tasks when context is insufficient. This is unlikely to conflict with other skills due to its specific focus on workspace decomposition and multi-agent delegation. | 3 / 3 |
Total | 9 / 12 Passed |
Implementation
22%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill provides a conceptual framework for workspace partitioning but falls short on actionability — it reads more like a design document than an executable skill. The two strategies (Summarize & Index vs Work Distribution) are a useful distinction, but neither is backed by concrete examples, templates, or step-by-step workflows with validation. The skill would benefit significantly from example subagent dispatch prompts, a concrete partitioning walkthrough, and explicit verification criteria.
Suggestions
Add a concrete, end-to-end example showing how to partition a real workspace (e.g., a monorepo with 150 files) into scoped subagent tasks, including the actual subagent prompt templates used for dispatch.
Create an explicit numbered workflow for each strategy (Summarize & Index, Work Distribution) with validation checkpoints — e.g., 'After scoping, verify no file appears in multiple scopes' and 'After subagent completion, verify all output files exist and cross-scope interfaces are consistent.'
Replace vague bullet points like 'Subagent: discoverer, explore, etc.' with concrete role definitions and example invocation patterns showing scope paths, goals, and output format.
Add a verification/conflict-resolution workflow with specific steps for handling cross-scope dependency failures and shared-interface conflicts, rather than just mentioning 'extra pass.'
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The content is moderately efficient but includes some redundancy and vague bullet points (e.g., 'Subagent: discoverer, explore, etc.' and 'Subagent: executor, engineer, etc.' add little value). The XML wrapper tags and role description are unnecessary overhead. However, it doesn't over-explain basic concepts. | 2 / 3 |
Actionability | The skill provides abstract descriptions and bullet-point lists rather than concrete, executable guidance. There are no code examples, no specific commands, no template prompts for subagent dispatching, and no concrete examples of how to partition a workspace or what a subagent invocation looks like. Phrases like 'Spawn subagents in parallel if possible' and 'Resolve shared-interface conflicts or changes with extra pass' are vague directions, not actionable instructions. | 1 / 3 |
Workflow Clarity | While there's a rough conceptual flow (detect task type → scope → dispatch → verify), the steps are not clearly sequenced with numbered workflows or explicit validation checkpoints. The verification step ('spawn another set of subagents to verify') is mentioned but lacks any detail on what to check or how to handle failures. For a skill involving coordinated multi-agent work with destructive file modifications, this lack of explicit sequencing and feedback loops is a significant gap. | 1 / 3 |
Progressive Disclosure | The content has some structure with sections for Summarization & Indexing vs Work Distribution, task type detection, and scoping. However, there are no bundle files and no references to external detailed guides. The content is somewhat monolithic with dense bullet lists that could benefit from being split into separate reference files for each strategy. References to other skills (reverse-engineering/SKILL.md, init-workspace-discovery/SKILL.md) are mentioned but not clearly navigable. | 2 / 3 |
Total | 6 / 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 |
|---|---|---|
frontmatter_unknown_keys | Unknown frontmatter key(s) found; consider removing or moving to metadata | Warning |
Total | 10 / 11 Passed | |
- Repository
- griddynamics/rosetta
- Commit
0f3002b
- 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.