CtrlK
BlogDocsLog inGet started
Tessl Logo

quasi-coder

Expert 10x engineer skill for interpreting and implementing code from shorthand, quasi-code, and natural language descriptions. Use when collaborators provide incomplete code snippets, pseudo-code, or descriptions with potential typos or incorrect terminology. Excels at translating non-technical or semi-technical descriptions into production-quality code.

64

1.00x
Quality

47%

Does it follow best practices?

Impact

96%

1.00x

Average score across 3 eval scenarios

SecuritybySnyk

Passed

No known issues

Optimize this skill with Tessl

npx tessl skill review --optimize ./skills/quasi-coder/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Discovery

67%

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 has good structural completeness with an explicit 'Use when' clause and a clear what/when split. However, it relies on somewhat jargon-heavy terms ('quasi-code', '10x engineer') rather than natural user language, and the capability description stays at a high level without enumerating specific concrete actions. The phrase 'Expert 10x engineer skill' is unnecessary fluff that doesn't aid skill selection.

Suggestions

Replace buzzwords like 'Expert 10x engineer skill' and 'quasi-code' with natural user terms like 'rough code drafts', 'code sketches', or 'broken code examples' that users would actually use in requests.

Add more specific concrete actions, e.g., 'Fixes syntax errors in rough code, infers intended logic from incomplete snippets, converts plain English task descriptions into working functions.'

DimensionReasoningScore

Specificity

Names the domain (interpreting shorthand/quasi-code) and some actions (translating descriptions into production-quality code), but doesn't list multiple concrete specific actions—it stays at a high level describing the general capability rather than enumerating distinct operations.

2 / 3

Completeness

Clearly answers both 'what' (interpreting and implementing code from shorthand, quasi-code, and natural language) and 'when' (when collaborators provide incomplete code snippets, pseudo-code, or descriptions with potential typos or incorrect terminology) with an explicit 'Use when' clause.

3 / 3

Trigger Term Quality

Includes some relevant keywords like 'pseudo-code', 'shorthand', 'incomplete code snippets', 'typos', and 'natural language descriptions', but misses common user phrasings like 'rough draft code', 'sketch of code', 'convert my idea to code', or 'fix my broken code'. The term 'quasi-code' is not something users naturally say.

2 / 3

Distinctiveness Conflict Risk

While it carves out a niche around interpreting imprecise code descriptions, it could easily overlap with general coding assistance skills or code generation skills. The boundary between 'translating natural language to code' and general code generation is fuzzy.

2 / 3

Total

9

/

12

Passed

Implementation

27%

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 verbose, spending most of its token budget explaining concepts Claude already understands (expert judgment, design patterns, handling varying expertise levels) rather than providing the specific, actionable shorthand-processing protocol. The shorthand notation system itself (()=> markers, start/end boundaries) is the genuinely useful novel content but is buried in extensive philosophical guidance. The entire skill could be reduced to ~50-80 lines covering the shorthand syntax, interpretation rules, and one example, which would be far more effective.

Suggestions

Cut the skill to ~50-80 lines: remove the Role section, Best Practices list, Resource Management section, and expertise level descriptions — Claude already knows how to be a good engineer. Focus on the novel shorthand notation protocol.

Move the Troubleshooting table, Advanced Usage examples, and Variables specification into separate bundle files (e.g., TROUBLESHOOTING.md, EXAMPLES.md) and reference them from a concise overview.

Replace the abstract confidence percentages (90%, 30%, 99%) with concrete decision criteria or examples showing what 'high confidence' vs 'low confidence' collaborator input actually looks like in practice.

Add a single concise workflow summary at the top (e.g., a numbered list of 4-5 steps) that serves as the primary reference, rather than scattering process steps across multiple sections.

DimensionReasoningScore

Conciseness

Extremely verbose at ~300+ lines. Explains concepts Claude already knows (what a 10x engineer is, what validation is, what design patterns are). The 'Role' section, 'Understanding Collaborator Expertise Levels', 'Best Practices', and 'Working with Tools and Reference Files' sections are largely restating things Claude inherently understands. The metaphor about architects reading sketches is repeated. Much of this could be condensed to under 50 lines.

1 / 3

Actionability

The shorthand notation system with `()=>` markers and start/end boundaries is concrete and specific. The example workflow showing input→output is helpful. However, much of the skill is abstract guidance ('apply expert judgment', 'compensate for erroneous descriptions') rather than executable instructions. The compensation rules use vague confidence percentages without clear criteria for determining them.

2 / 3

Workflow Clarity

The Example Workflow section provides a clear 3-step process (Assess → Interpret → Implement) with a concrete example. However, the overall skill lacks a concise, unified workflow for processing shorthand. The interpretation process in the Shorthand Interpretation section lists steps but doesn't include validation checkpoints or feedback loops for verifying the implementation matches intent.

2 / 3

Progressive Disclosure

Monolithic wall of text with no references to external files. All content is inline despite many sections (Resource Management, Troubleshooting, Advanced Usage, Variables spec) that could be split into separate reference files. No bundle files exist to support progressive disclosure. The content would benefit greatly from being split into a concise overview with references to detailed guides.

1 / 3

Total

6

/

12

Passed

Validation

100%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation11 / 11 Passed

Validation for skill structure

No warnings or errors.

Repository
github/awesome-copilot
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.