inquire
Infer context insufficiency before execution. Surfaces uncertainties through information-gain prioritized inquiry when AI infers areas of context insufficiency, producing informed execution. Type: (ContextInsufficient, AI, INQUIRE, Prospect) → InformedExecution. Alias: Aitesis(αἴτησις).
33
17%
Does it follow best practices?
Impact
Pending
No eval scenarios have been run
Risky
Do not use without reviewing
Optimize this skill with Tessl
npx tessl skill review --optimize ./aitesis/skills/inquire/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 over-engineered with academic jargon, type signatures, and a Greek alias, making it nearly impenetrable for skill selection purposes. It fails to communicate concrete actions in plain language and lacks any natural trigger terms a user would employ. The core concept—asking clarifying questions when information is missing—is buried under unnecessary abstraction.
Suggestions
Rewrite in plain language describing concrete actions, e.g., 'Asks targeted clarifying questions when a user request lacks sufficient context to execute properly.'
Add an explicit 'Use when...' clause with natural trigger terms, e.g., 'Use when a user request is ambiguous, underspecified, or missing key details needed to proceed.'
Remove the type signature notation and Greek alias, which add no value for skill selection and obscure the description's purpose.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description uses highly abstract, academic language ('context insufficiency', 'information-gain prioritized inquiry', 'InformedExecution') without listing any concrete actions a user would recognize. No specific tasks or domains are mentioned. | 1 / 3 |
Completeness | The 'what' is buried in abstract language and the 'when' is only vaguely implied ('when AI infers areas of context insufficiency'). There is no explicit 'Use when...' clause with actionable triggers, and the actual purpose remains unclear. | 1 / 3 |
Trigger Term Quality | The description contains no natural keywords a user would say. Terms like 'ContextInsufficient', 'Aitesis(αἴτησις)', and 'Prospect' are technical jargon and type signatures that no user would naturally use in a request. | 1 / 3 |
Distinctiveness Conflict Risk | The concept of asking clarifying questions when context is insufficient is somewhat distinct, but the description is so abstract that it's hard to know when it should trigger versus any other skill that might involve asking questions or gathering requirements. | 2 / 3 |
Total | 5 / 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 an extremely dense formal protocol specification that prioritizes mathematical rigor over practical usability. While the underlying concept (inferring context insufficiency before execution) is valuable, the implementation drowns actionable guidance in formal type theory, category theory notation (Grothendieck fibrations), and exhaustive edge-case enumeration. The skill would benefit enormously from a concise overview with practical examples, relegating formal specifications to reference files.
Suggestions
Reduce the main SKILL.md to a concise overview (~100 lines) covering the core concept, phase workflow summary, and 1-2 concrete examples of what an Aitesis interaction looks like in practice, moving formal type definitions to a TYPES.md reference file.
Add concrete examples showing actual Aitesis interactions — e.g., a sample prospect, identified uncertainties, Phase 2 surfacing output, and user response flow — so Claude can pattern-match on real behavior rather than interpreting abstract notation.
Move the detailed rules (22 rules), UX safeguards table, formal type system, and mode state specification into separate referenced files (e.g., RULES.md, TYPES.md, SAFEGUARDS.md) with clear one-level-deep navigation links.
Remove or drastically compress the formal mathematical notation (fibration structure, dependent sums, subset types) — Claude does not need category theory to understand 'classify uncertainties by dimension and verifiability, then route to appropriate resolution channel.'
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | This skill is extremely verbose (~600+ lines) and contains extensive formal type theory notation, mathematical definitions, and abstract protocol specifications that Claude does not need explained at this level of detail. The content reads like an academic paper rather than actionable instructions, with massive amounts of formal notation (Grothendieck fibrations, dependent sums, fiber types) that add token cost without proportional clarity gain. | 1 / 3 |
Actionability | The skill does provide concrete phase-by-phase workflows and specific rules for behavior, but the guidance is buried under layers of formal notation. There are no executable code examples, no concrete input/output demonstrations, and the instructions are heavily abstract rather than showing what an actual Aitesis interaction looks like in practice. | 2 / 3 |
Workflow Clarity | The multi-phase workflow (Phase 0→1→2→3) is clearly sequenced with explicit transitions and validation checkpoints (escape conditions, staleness guards, observation constraints). However, the workflow is so complex and densely specified that following it in practice would be extremely difficult. The formal notation obscures rather than clarifies the actual steps, and backward arcs and reclassification paths create a labyrinthine flow. | 2 / 3 |
Progressive Disclosure | The entire specification is monolithic — all formal type definitions, phase transitions, rules, UX safeguards, composition details, and mode state are inlined in a single massive document. There are no references to separate files for detailed specifications (e.g., the type system, the rules, the examples could each be separate files). The content that should be split across reference documents is all presented inline. | 1 / 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 |
|---|---|---|
skill_md_line_count | SKILL.md is long (525 lines); consider splitting into references/ and linking | Warning |
Total | 10 / 11 Passed | |
- Repository
- jongwony/epistemic-protocols
- Commit
a3403de
- 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.