CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

plan-deck-sample

海岚旅行《方案 PPT 范例》(HL-MKT-SMP-01):一份往期已交付、已脱敏的方案 PPT,留作参考,看合格的方案写成什么样。范例里的客人、日期、价格、事实与来源都属于那位客人,不得搬进别的客人的方案。

61

Quality

77%

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 ./experimental/simulation/scenarios/travel_agency/materials/plan-deck-sample/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

92%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 an exemplary short instruction skill: lean, specific, and unambiguous about how the sample may and may not be used, with proper staleness handling for versioned references. Its one real defect is that the referenced sample.pptx is missing from the reviewed bundle, leaving the skill's core artifact unreachable and the reference dangling.

Suggestions

Ship sample.pptx inside the skill directory (or correct the path in SKILL.md) so the single referenced artifact actually exists — this alone would lift progressive_disclosure to the top of the scale.

Briefly indicate where the cross-referenced governing documents (HL-MKT-SPEC-04, HL-MKT-GUIDE-07, HL-FIN-PRC-06) live, so the pointers resolve for a reader outside the marketing team.

DimensionReasoningScore

Conciseness

The body is lean with zero concept-explanation padding: every line states the file's identity, governing documents, contents, or usage rules. The one version-sensitive citation ("价目表 HL-FIN-PRC-06 V3.1") is correctly handled with the staleness caveat "规则有变化时,以现行材料为准", so no token is wasted or misleading. Not below 5 — nothing could be trimmed without losing information.

5 / 5

Actionability

Although instruction-only (no code, which is appropriate here), the guidance is fully specific: exact file paths ("本目录下的 sample.pptx", "plan-deck-template 目录下的 template.pptx"), document IDs with versions, and an enumerated list of fields that must never be copied (称呼、日期、人数、目的地、价格、报价编号、付款金额与日期、交通时长、酒店、门票和来源). This matches the top anchor for instruction skills — concrete and unambiguous.

5 / 5

Workflow Clarity

This is a simple, single-purpose reference skill and its single action is unambiguous: study the sample's writing and layout only, copy nothing client-specific, and build new decks from template.pptx instead. No destructive or batch operations are involved, so no validation checkpoints are required, and the simple-skill provision applies.

5 / 5

Progressive Disclosure

The structure itself is ideal — a short overview clearly signaling a single one-level-deep reference ("本目录下的 sample.pptx") — but the referenced core file is absent from the bundle as reviewed (the directory contains only SKILL.md, config.json, and rubrics). Since navigation to the skill's sole content artifact fails, this is more than a minor organization gap, placing it at 3 rather than 4; it is not lower because the reference is clearly signaled and correctly scoped.

3 / 5

Total

18

/

20

Passed

Description

50%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 clearly states what the skill is (a de-identified, previously delivered proposal PPT kept as a quality reference) and sets an important data-leakage boundary, but it omits any explicit 'when to use' trigger and does not separate itself from the sibling template/spec skills. Trigger-term coverage is adequate but thin on synonyms.

Suggestions

Add an explicit trigger clause, e.g. 「当需要参考往期已交付方案、对照合格方案应写成什么样时使用」— this would raise completeness above the cap of 3.

Include natural trigger synonyms users might say: 样例、案例、往期方案、参考案例, alongside the existing 范例/参考.

Distinguish the skill from its sibling template skill in the description, e.g. 「制作新方案请使用 plan-deck-template;本技能仅用于对照参考」, to reduce conflict risk within the proposal-PPT skill family.

DimensionReasoningScore

Specificity

The description names its domain concretely ("一份往期已交付、已脱敏的方案 PPT,留作参考,看合格的方案写成什么样") and states one clear function (serve as a reference exemplar) plus a data-handling constraint ("不得搬进别的客人的方案"). It stays at 3 rather than 4 because it lists only this single purpose with no broader set of concrete actions, and above 2 because it is far more informative than a bare domain label like "Processes PDF files".

3 / 5

Completeness

The 'what' is clear (a delivered, de-identified sample proposal PPT kept as a reference), but there is no 'Use when...' clause or equivalent explicit trigger guidance — only the weakly implied "留作参考". Per the judging guideline, a missing explicit trigger caps completeness at 3.

3 / 5

Trigger Term Quality

Natural terms are present — "方案 PPT", "范例", "参考", "合格" — which a user asking to see a past proposal example would plausibly say. It sits at 3 rather than 4 because common variations and synonyms are missing (e.g. 样例, 案例, 往期方案, 参考案例) and there is no explicit trigger phrasing to anchor discovery.

3 / 5

Distinctiveness Conflict Risk

The unique document ID (HL-MKT-SMP-01) and the sample-vs-template framing give it a niche, but the description never distinguishes it from closely related sibling skills in the same proposal-PPT family (the template, spec, and guide it references in the body). A user mentioning "方案 PPT" could reasonably match several of these, so overlap risk is more than minor — 3 rather than 4.

3 / 5

Total

12

/

20

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.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository

EverMind-AI/Raven

Path

experimental/simulation/scenarios/travel_agency/materials/plan-deck-sample/SKILL.md

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.