deep-interview
Socratic deep interview with mathematical ambiguity gating before explicit execution approval
32
27%
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 ./skills/deep-interview/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 abstract and jargon-laden, failing to communicate concrete actions or provide natural trigger terms. It reads more like an academic concept label than a functional skill description. Without clear 'what' and 'when' guidance, Claude would struggle to appropriately select this skill from a list.
Suggestions
Replace abstract terminology with concrete actions, e.g., 'Conducts structured interviews by asking clarifying questions to resolve ambiguous requirements before proceeding with task execution.'
Add an explicit 'Use when...' clause with natural trigger terms, e.g., 'Use when the user's request is ambiguous, underspecified, or when confirming requirements before executing complex tasks.'
Remove or define jargon like 'mathematical ambiguity gating' — either explain what it means in plain language or replace with terms users would naturally use such as 'requirement clarification' or 'disambiguation'.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description uses abstract, jargon-heavy language ('Socratic deep interview', 'mathematical ambiguity gating', 'explicit execution approval') without listing any concrete actions the skill performs. No specific capabilities like 'asks clarifying questions' or 'validates requirements' are mentioned. | 1 / 3 |
Completeness | The description fails to clearly answer either 'what does this do' or 'when should Claude use it'. There is no 'Use when...' clause or equivalent trigger guidance, and the 'what' is expressed in abstract terms that don't convey practical functionality. | 1 / 3 |
Trigger Term Quality | The terms used are highly technical and unlikely to match natural user language. No user would naturally say 'mathematical ambiguity gating' or 'Socratic deep interview' when requesting help. There are no common, natural keywords present. | 1 / 3 |
Distinctiveness Conflict Risk | The unusual terminology ('Socratic deep interview', 'mathematical ambiguity gating') is distinctive enough that it's unlikely to conflict with common skills, but the lack of clarity about what it actually does means it could either never trigger or trigger inappropriately. | 2 / 3 |
Total | 5 / 12 Passed |
Implementation
47%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a deeply thought-out workflow skill with excellent sequencing, validation gates, and explicit approval checkpoints that prevent premature execution. Its primary weakness is extreme verbosity — the same concepts (threshold resolution, topology handling, execution gating) are restated multiple times across sections, and the Advanced section largely duplicates the Steps section in tabular form. The skill would benefit significantly from extracting the spec template, scoring prompts, and configuration reference into separate bundle files.
Suggestions
Reduce verbosity by at least 40%: remove the Advanced section's redundant tables (Brownfield vs Greenfield Weights, Challenge Agent Modes, Ambiguity Score Interpretation) that duplicate information already in the Steps, or move them to a separate REFERENCE.md bundle file.
Extract the spec template (Phase 4's markdown block) and the scoring prompt (Step 2c's prompt block) into separate bundle files like `SPEC_TEMPLATE.md` and `SCORING_PROMPT.md` to reduce the main skill's token footprint.
Consolidate the threshold resolution logic — it's described in Phase 0, re-verified in Step 3.5, referenced in the announcement, and restated in the checklist. A single authoritative section with brief back-references would suffice.
Remove explanatory rationale paragraphs (e.g., 'Why_This_Exists', 'Why 3 stages?') that explain motivation rather than instruct — Claude doesn't need to understand why the skill was designed this way to execute it correctly.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | This skill is extremely verbose at ~600+ lines. It explains concepts Claude already understands (what Socratic questioning is, why clarity matters, what PDFs are in analogous terms), includes extensive pipeline diagrams, redundant checklists, and repeats the threshold resolution logic multiple times across phases. The Phase 0 threshold resolution alone could be 5 lines but spans ~30. The Advanced section largely restates what was already covered in the Steps section. | 1 / 3 |
Actionability | The skill provides concrete JSON state schemas, specific scoring formulas, table formats for reporting, and clear prompt templates for scoring. However, much of the guidance is procedural description rather than executable code — there are no actual executable scripts, the scoring prompt is a template rather than a runnable artifact, and the Skill() invocations are described but not demonstrated with complete parameters. The ambiguity scoring formula and dimension weights are concrete and actionable. | 2 / 3 |
Workflow Clarity | The multi-phase workflow is exceptionally well-sequenced: Phase 0 (blocking threshold resolution) → Phase 1 (initialization with brownfield detection) → Round 0 (topology gate) → Phase 2 (interview loop with explicit sub-steps 2a-2f) → Phase 3 (challenge agents at specific thresholds) → Phase 4 (spec crystallization) → Phase 5 (approval-gated execution bridge). Validation checkpoints are explicit throughout — ambiguity scoring after every round, topology confirmation before scoring, threshold verification before proceeding, and a 3-stage approval gate preventing premature execution. | 3 / 3 |
Progressive Disclosure | The skill references external files like `docs/company-context-interface.md` and other skills via `Skill()` invocations, which is good. However, the SKILL.md itself is monolithic — the Advanced section repeats tables and pipeline diagrams already present in Steps, and there are no bundle files to offload the extensive configuration reference, scoring rubric details, or spec template to separate documents. The content that could be in referenced files (spec template, scoring prompt, configuration schema) is all inline. | 2 / 3 |
Total | 8 / 12 Passed |
Validation
81%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 9 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
skill_md_line_count | SKILL.md is long (803 lines); consider splitting into references/ and linking | Warning |
frontmatter_unknown_keys | Unknown frontmatter key(s) found; consider removing or moving to metadata | Warning |
Total | 9 / 11 Passed | |
- Repository
- Yeachan-Heo/oh-my-claudecode
- Commit
3e94567
- 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.