deep-interview

Socratic deep interview with mathematical ambiguity gating before explicit execution approval

Quality

64%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./skills/deep-interview/SKILL.md

Quality

Content

70%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is highly actionable and exceptionally well-sequenced with explicit gates, feedback loops, and a checklist. Its weaknesses are token efficiency (heavy duplication of the pipeline/threshold content across sections) and progressive disclosure (an 800-line monolith with no reference-file split and a dangling external reference).

Suggestions

Deduplicate the approval-gated pipeline content — currently described in the Purpose, Execution_Policy, Phase 5 ASCII diagram, and the Advanced 'Approval-Gated Pipeline' section — into a single canonical statement; this alone would remove a large padded block and lift conciseness.

Move the verbatim spec template, the scoring prompt, and the Advanced configuration reference into separate files under references/ (e.g. references/spec-template.md, references/scoring-prompt.md) and link to them from SKILL.md, shrinking the monolith and improving progressive disclosure.

Fix or remove the dangling 'See docs/company-context-interface.md' reference (no such file exists in the bundle), either by creating that reference file or by inlining the companyContext interface contract.

Dimension	Reasoning	Score
Conciseness	At ~800 lines / 48KB the body is noticeably verbose: the approval-gated pipeline is restated in the Purpose, Execution_Policy, Phase 5 ASCII diagram, and a long Advanced section; the Phase 0 threshold marker, artifact-path discipline, and 'pending approval' phrasing are repeated throughout — matching the 'noticeably verbose; several padded sections' anchor rather than the mostly-efficient anchor at 3.	2 / 5
Actionability	The body provides copy-paste-ready verbatim templates — the full state JSON schema, the scoring prompt, the spec markdown structure, the announcement text, and exact Skill() invocations with flags — and covers common cases (greenfield/brownfield, single/multi-component, early exit), matching the 'fully executable; copy-paste ready; covers common cases' anchor.	5 / 5
Workflow Clarity	The process is explicitly phased (Phase 0 → Phase 1 → Round 0 → Phase 2 loop → Phase 3 → Phase 4 → Phase 5) with numbered steps, hard validation gates (Phase 0 blocking prerequisite, ambiguity ≤ threshold, Round 0 topology lock), a score→target→re-score feedback loop, and a final checklist, matching the 'clear sequence with explicit validation, feedback loops, and checklists' anchor; the destructive-cap does not apply since the skill is a non-mutating requirements gate.	5 / 5
Progressive Disclosure	No bundle files exist and the body is a single 800-line monolith with bulk content (spec template, scoring prompt, config reference, examples) inlined rather than split into reference files; the one in-body reference 'docs/company-context-interface.md' is dangling (no such file), so this fits the 'some structure but content that should be separate is inline' anchor better than the well-split anchor at 4.	3 / 5
	Total	15 / 20 Passed

Description

58%Weight 40%Scale 1-5

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 concretely names the skill's core mechanisms and occupies a distinct niche, but it lacks any explicit 'Use when...' trigger guidance and omits the natural trigger phrases users would actually say. Specificity is solid; completeness and trigger-term coverage are the limiting dimensions.

Suggestions

Add an explicit 'Use when...' clause with concrete triggers — e.g., 'Use when the user has a vague idea, says "interview me" / "ask me everything" / "don't assume", or wants requirements gathered before execution' — to lift completeness above the 3 cap.

Include the natural trigger synonyms users actually say ('interview me', 'ask me everything', 'clarify my idea', 'requirements gathering') instead of the technical 'mathematical ambiguity gating' phrasing to improve trigger-term coverage toward 4-5.

Optionally reframe to lead with the action ('Conducts a Socratic deep interview that ...') and append the trigger clause, keeping the distinctiveness while making the when-to-use explicit.

Dimension	Reasoning	Score
Specificity	Phrases 'Socratic deep interview', 'mathematical ambiguity gating', and 'explicit execution approval' name several concrete mechanisms (interview, ambiguity scoring/gating, approval gate), fitting the 'lists several specific actions; minor gaps' anchor better than the 1-2-action anchor at 3.	4 / 5
Completeness	The 'what' is clear (Socratic interview + ambiguity gating + approval gate), but there is no 'Use when...' clause or equivalent trigger guidance, so per the rubric cap completeness stays at 3 rather than the 'both what and when' anchor at 4.	3 / 5
Trigger Term Quality	'Socratic deep interview' supplies natural keywords, but the description omits the common phrasings users actually say ('interview me', 'ask me everything', 'clarify my idea', 'requirements gathering') and 'mathematical ambiguity gating' leans technical, matching the 'some relevant keywords but missing common variations' anchor.	3 / 5
Distinctiveness Conflict Risk	'Socratic deep interview with mathematical ambiguity gating' is a fairly distinct niche with low conflict risk, but without a Use-when trigger it has minor overlap risk with closely related planning skills (omc-plan, autopilot), fitting the 'mostly distinct; minor overlap' anchor at 4 rather than the minimal-conflict anchor at 5.	4 / 5
	Total	14 / 20 Passed

Validation

87%

Warnings & errors only

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

Validation — 14 / 16 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	14 / 16 Passed

Repository: Yeachan-Heo/oh-my-claudecode
Path: skills/deep-interview/SKILL.md
Commit: 5aa678c

Reviewed: 8 days ago

Table of Contents

Discovery Implementation Validation

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.