CtrlK
BlogDocsLog inGet started
Tessl Logo

pua-ja

お前のAIを詰めろ。日本企業の詰め文化と体系的デバッグ方法論で全ての手段を尽くさせる。トリガー条件:(1) タスク失敗2回以上または同じアプローチの微調整ループ; (2)「解決できません」と言おうとする・手動対応を推奨・未検証で環境を原因帰属; (3) 受け身——検索しない・ソースを読まない・指示待ち; (4) ユーザーの不満:'もっと頑張れ'、'なんでまた失敗したの'、'なんとかしろ'。全タスクタイプ適用。初回失敗や既知修正の実行中はトリガーしない。

57

Quality

65%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./codebuddy/pua-ja/SKILL.md

The canonical home for this skill is pua-ja in tanweai/pua

SKILL.md
Quality
Evals
Security

Quality

Content

62%

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

The skill delivers strong, concrete operational methodology with clear sequenced workflows and validation checklists. Its principal weakness is extreme verbosity from the persona and motivational packaging, and the lack of any file-based progressive disclosure for a document of this length.

Suggestions

Cut the theatrical persona monologues and corporate-flavor speeches substantially; condense the motivational layer so the operational methodology dominates the token budget.

Split the corporate-flavor packs and the Agent Team integration into separate reference files (e.g. flavors.md, agent-team.md) referenced one level deep from SKILL.md to enable progressive disclosure.

Add one or two fully worked examples applying the 5-step methodology end-to-end to ground the abstract steps in concrete demonstration.

DimensionReasoningScore

Conciseness

At ~370 lines the body is heavily padded with theatrical persona monologues, corporate-flavor speeches (Toyota/Dentsu/Rakuten/etc.), and motivational rhetoric that re-expresses motivation Claude already understands rather than operational guidance.

1 / 3

Actionability

Despite being instruction-only, the guidance is concrete and specific: read 50 lines around the error, list 3 essentially different hypotheses, isolate a minimal repro, and verify with build/test/curl, with per-task-type equivalents in the checklists.

3 / 3

Workflow Clarity

A clearly sequenced 5-step methodology with explicit ordering, an L1-L4 pressure-escalation table with forced actions, and the 7-item and proactivity checklists provide explicit validation checkpoints and feedback loops for complex processes.

3 / 3

Progressive Disclosure

The document is a single ~370-line monolithic file with no bundle/reference files; the corporate-flavor packs and Agent Team section are content that arguably should be split out, though internal section organization is clear.

2 / 3

Total

9

/

12

Passed

Description

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 is well-formed in third person and clearly answers both what and when, with explicit trigger guidance. Its main weakness is that the trigger terms are behavioral conditions rather than natural user-facing keywords, limiting distinctiveness and trigger-term quality.

Suggestions

Add natural user-facing trigger phrases a person would actually say (e.g. mentions of being stuck, repeated failed attempts, or dissatisfaction with output quality) rather than only internal behavioral states.

Expand the action list beyond two high-level actions to concrete specific capabilities the coaching mode performs.

Sharpen distinctiveness by tying triggers to this skill's unique coaching/pressure framing rather than generic verification states that overlap with debugging skills.

DimensionReasoningScore

Specificity

Names a domain ('生産性コーチングモード') and two actions ('構造化トラブルシューティング', '証拠ベースの完了確認を促す'), but the action list is not comprehensive enough for the top anchor.

2 / 3

Completeness

It states what the skill does and gives an explicit 'when' clause ('…のときに') listing five trigger conditions, satisfying both the 'what' and 'when' with explicit triggers.

3 / 3

Trigger Term Quality

The triggers ('明示的な依頼', '反復失敗', '受け身', '検証不足', '品質不満') are internal behavioral failure signals rather than natural vocabulary a user would utter to invoke the skill, so coverage is partial.

2 / 3

Distinctiveness Conflict Risk

The behavioral triggers (passivity, lack of verification, quality dissatisfaction) are broadly applicable and could overlap with general debugging/verification skills, despite the distinct coaching-mode framing.

2 / 3

Total

9

/

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.

Validation16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
tanweai/pua
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.