agent-response-style

Default response behavior for AI coding agents: professional, factual, neutral tone with calibrated critical evaluation. Baseline for every task: implementation, code review, debugging, planning, research, evaluation, Q&A.

Quality

61%

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 ./.claude/skills/agent-response-style/SKILL.md

Quality

Content

67%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 well-structured and gives concrete, actionable direction for an instruction-only skill, with a clear numbered workflow and sensible sectioning; its main weakness is repetition of the voice distinction and padded imperative prose in the teacher section.

Suggestions

State the peer-review-vs-teaching voice distinction once in 'Project context' and reference it from the other sections instead of re-explaining it in Baseline Behavior and the teacher section.

Tighten the teacher section by collapsing the repeated 'Make sure they understand...' imperatives into a single concise checklist.

Consider moving the long 'Verbatim Directives' block into a references file and keeping a short summary in SKILL.md to reduce inline length.

Dimension	Reasoning	Score
Conciseness	The body is mostly directive and on-topic, but the peer-review-vs-teaching distinction is restated in three sections, and the teacher section repeats imperatives ('Make sure they understand the why. Make sure they understand the what, and the how.'), so it could be tightened without losing meaning.	3 / 5
Actionability	For an instruction-only skill the guidance is concrete and specific — 'present 2 to 4 credible alternatives', 'name the strongest counter-argument explicitly before recommending', 'Quiz them with open-ended or multiple-choice questions with AskUserQuestion' — with a numbered 5-step workflow, leaving only minor gaps.	4 / 5
Workflow Clarity	The 'Suggested Response Workflow' gives a clear 5-step sequence (clarify, compare, evaluate each, state distinguishing evidence, ask framing questions), and the critical-evaluation flow has implicit checkpoints, though error/feedback loops are not relevant to this non-destructive skill.	4 / 5
Progressive Disclosure	Content is well-organized under clearly labeled sections (Purpose, Baseline Behavior, Verbatim Directives, Workflow, Humanize, Transparency, Teacher) with no nested references and only a single one-level reference to the external humanizer skill; at ~90 lines some inlined material could be trimmed but nothing clearly belongs in a separate file.	4 / 5
	Total	15 / 20 Passed

Description

55%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 is clear about what the skill does and asserts an always-on baseline scope, but that breadth creates high conflict risk and the trigger framing relies on stance language users would not naturally utter.

Suggestions

Add an explicit 'Use when...' clause naming concrete trigger moments (e.g. 'Use when reviewing a design, evaluating a proposal, or choosing between approaches') instead of relying on 'Baseline for every task'.

Replace abstract stance words ('calibrated critical evaluation') with one or two natural user phrasings ('push back on my plan', 'compare options') to improve trigger-term quality.

Narrow the scope claim so the skill is distinguishable from general coding skills and less likely to co-fire with everything.

Dimension	Reasoning	Score
Specificity	The description names the domain ('Default response behavior for AI coding agents') and lists several concrete attributes ('professional, factual, neutral tone with calibrated critical evaluation') plus a task list, but these are task categories and tone adjectives rather than the concrete executable actions the anchor rewards.	3 / 5
Completeness	It states a clear 'what' (default response tone and critical-evaluation stance) and an explicit 'when' ('Baseline for every task: implementation, code review, debugging, planning, research, evaluation, Q&A'), but the 'when' is a broad 'every task' baseline rather than a sharply targeted trigger phrase, keeping it just below a 5.	4 / 5
Trigger Term Quality	It includes natural task-type terms users do say ('code review', 'debugging', 'planning', 'research', 'Q&A'), but the framing is around a tone/stance ('calibrated critical evaluation') that users rarely voice, so coverage of natural phrasings is partial.	3 / 5
Distinctiveness Conflict Risk	By design it is a 'Baseline for every task' that applies alongside any other skill, so it has high overlap/conflict risk with virtually every coding skill rather than a distinct, narrowly-triggered niche.	2 / 5
	Total	12 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: SRombauts/SQLiteCpp
Path: .claude/skills/agent-response-style/SKILL.md
Commit: 672a1e0

Reviewed: 5 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.