Content & Media Creation · dontbesilent2025/dbskill

dbs-resonate

Name: dbs-resonate
Author: dontbesilent2025/dbskill

Diagnoses resonance mechanisms, problem locations, and revision directions using communication and social psychology theories tailored to the draft. Ideal when users worry about low traffic, failing to resonate with audiences, low completion rates, or are unsure whether content is ready to publish.

L1 Basic · Source registration onlyV1 Commit-PinnedDual-pass Audit

Author dontbesilent2025GitHub dontbesilent2025/dbskill ↗Stars 10121License CC BY-NC 4.0（仓库根 LICENSE：Attribution-NonCommercial 4.0 International (CC BY-NC 4.0)；署名 + 禁止商业使用；Copyright (c) 2026 dontbesilent）Commit 8b8e33f1ec

Control levels describe skill capabilities and suggested control measures; they do not represent quality or safety certifications, nor do they dictate usability. Usage decisions rest entirely with users and administrators. Different agent hosts apply varying constraints; do not assume hosts provide automatic protection. Third-party skill copyrights belong to their original authors. Use at your own risk.

1How It Works & Mechanism

A pure prompt-orchestrated theoretical diagnostic engine: The knowledge base is an inlined table in SKILL.md featuring 5 primary theories + 3 conditional theories (including authors, years, and DOIs), breaking 'resonance' down into a step-by-step screening checklist of mechanisms.

skills/dbs-resonate/SKILL.md

| 社会认同与自我分类 | Tajfel、Turner，1979；Turner 等，1987 |

Theoretical entries include literature anchors (e.g., Cognitive Appraisal Theory with DOI: 10.1037/0022-3514.48.4.813), ensuring theoretical citations can be externally verified rather than relying on free association.

The diagnostic workflow follows a strict 5-step sequence: 'Define claim → Select mechanism → Locate problem → Prescribe revisions', avoiding premature rewriting suggestions.

skills/dbs-resonate/SKILL.md

### Step 2：确定核心主张和目标动作

Step 2 requires answering 'If only one claim can be kept, which one should it be?' and 'What primary action does the draft expect the audience to take?'.

Core Discipline 1: Retain only one core claim per draft; multiple parallel claims are flagged as 'diluted core'.

skills/dbs-resonate/SKILL.md

### 1. 一篇文稿只保留一个核心主张

All other viewpoints must be demoted to evidence, case studies, explanations, or action advice.

Core Discipline 2: Select only one primary theory per diagnosis; secondary theories must meet three strict conditions (covering a new mechanism + explicit trigger evidence in the original text + removing it would lose a key recommendation).

skills/dbs-resonate/SKILL.md

1. 它解释主理论没有覆盖的新机制；

Conditional theories (Spiral of Silence / Cognitive Dissonance / Two-Step Flow) must satisfy their specific trigger criteria and strictly forbid over-generalization: 'Never generalize conditional theories into everyday resonance tags'.

Core Discipline 3: Every conclusion must tie back to specific sentences in the source text—identifying which sentence activates which mechanism, where the gap occurs, and how to revise it.

skills/dbs-resonate/SKILL.md

每条结论都要回答：哪句话激活了什么机制，缺失发生在哪句话，应该怎样改。

Step 4 classifies mechanism status into three levels: Effective / Weak / Ineffective, and outlines five possible missing conditions (identity, attribution, appraisal path, utility, sharing rationale).

Core Discipline 4: Decouple 'draft capability' from 'platform outcomes', explicitly listing three out-of-scope commitments (predicting metrics, promising virality, analyzing already viral content).

skills/dbs-resonate/SKILL.md

预测完播率、点击率、推荐量等平台指标；

Clear routing follows the boundary: 'To analyze why already verified content went viral, preserve materials and refer back to the /dbs entry point'.

Output format is standardized into a 6-section diagnostic report with icons (Core Mechanism Review, Theoretical Diagnosis, Concrete Revisions, etc.), with revisions categorized into four specific actions: Delete, Compress, Reinforce, or Keep.

skills/dbs-resonate/SKILL.md

删除：[具体内容和原因]

The final line of the template requires providing 'Revised Skeleton: One-sentence single content structure', converging diagnosis into an actionable rewriting direction.

Includes 5 reusable common problem patterns (ambiguous target identity, framework lacking causal bridge, emotion labels only, unclear utility, insufficient sharing incentive) acting as diagnostic accelerators.

skills/dbs-resonate/SKILL.md

### 模式 3：只有情绪标签

Each pattern specifies 'what to fix': e.g., ambiguous identity → add shared predicament, group boundaries, and concrete interests.

2Core Capabilities

01Claim audit: List all claims and evaluate whether the core is clear / diluted / unclear

02Target action identification (comprehend / agree / feel relieved / bookmark / comment / share / take action)

03Sequential screening of 5 mechanisms (Identity → Framing → Emotional Appraisal → Content Utility → Online Sharing) and selecting a single primary theory

04Mechanism implementation assessment (Effective / Weak / Ineffective + sentence-level precision)

05Actionable revision plan: What to delete, what to compress, which sentence to reinforce, what to keep intact, and revised skeleton

06Conditional theory gating: Only permit Spiral of Silence, Cognitive Dissonance, or Two-Step Flow when trigger conditions are met

077-item self-check checklist, turning diagnostic quality control into actionable checkboxes

08Boundary enforcement: Refuse platform metric predictions, refuse virality guarantees, route published content back to /dbs

4Recommended Governance & Evidence

Suggested Controls for Safe Operation

For Agent Runtime

Can be auto-loaded

For Administrators

Evidence & Triggers (File Paths & Excerpts)

SKILL.md: Pure prompt instructions and rule text, without script execution or external side effects

Audit Notes: The evaluation target is the skill's own directory: contains only SKILL.md (inlined theory table) and an interface metadata yaml, with no scripts, network endpoints, credential reads, or disk writes—making it purely instruction-based. The considerations are cognitive: it provides 'textual evidence judgment of mechanisms' rather than traffic predictions, and theoretical selection is performed by the model, which might occasionally misjudge the primary theory; the skill mitigates this through single-theory discipline, triple-gated auxiliary theory thresholds, and self-check checklists.

5Second-Pass Verification

[ok] 外部资源/工具：是否引用了 SKILL.md 未内联的外部文献或服务 — 理论仅以「作者+年份+DOI」文字形式内联，无链接抓取、无参考文献文件、无 API。DOI 只是可核对标识，不构成网络调用。external_deps 空数组成立。
[ok] 安全结论：是否存在文件读写遗漏 — 全文无读取文件指令（不要求读用户草稿文件，用户直接贴进对话即可），无落盘要求；输出是对话内的 markdown 诊断报告。
[ok] 实现原理是否成立：理论机制是否真能落到「改哪一句」 — Step 4 要求「说明问题出现在哪一句，并写清缺少身份、归因、评价路径、内容用途或传播理由中的哪一项」；Step 5 与输出模板的【具体改法】要求给出具体原句与改法。机制到句子的映射是设计内的强制项，不是可选建议。
[ok] 理论表完整性：条件理论是否真的「只在条件下使用」 — 条件理论表三行各带触发条件，且正文另加禁令「禁止把条件理论泛化成日常共鸣标签」；Step 3 再次要求「条件理论必须满足对应触发条件」。三处要求一致，无冲突。
[unlocatable] 证据定位：仓库级元数据（stars/last_push）与 installs — stars=10121、last_push=2026-09-07 来自站点池 data/pool.json（GitHub API 采集），pin 源码内无法逐字定位；installs 未采集记 null。license 可在仓库根 LICENSE 逐字命中。

6Verdict & Scenarios

Translates vague anxiety like 'content gets no traffic' into actionable mechanism checks: Is the claim singular? Is the mechanism realized? Which link is missing? Which sentence should be modified or cut?

Accountable theoretical citations: 5 primary theories + 3 conditional theories all include sources and DOIs with explicit 'usage boundaries', preventing misuse of theories as arbitrary appeals to authority

Exceptionally thorough anti-hallucination and anti-overpromising controls: Prohibits predicting platform metrics, promising virality, over-generalizing conditional theories, and parallel multiple claims

Immediately actionable outputs: Delete/Compress/Reinforce/Keep action types + a one-sentence new skeleton allow direct implementation in revisions

Comes with five high-frequency problem patterns that even beginners can easily apply

Best for: Ideal for content creators conducting a pre-publication mechanism checkup: long-form articles, official accounts, spoken video scripts, and RED (Xiaohongshu) posts—especially when 'metrics are consistently poor but the root cause is unclear'. Also suitable as a writing checklist (5 patterns + 7 self-checks) for team editorial reviews.

Not for: Not suitable for platform metric predictions or ad spend advice (explicitly out of scope); not suitable for post-mortem analysis of published viral content (refer back to /dbs for diagnostic intake); not suitable for evaluating purely utility/search-oriented content (the table states group identity theory may not apply); not suitable for overall strategic planning of multi-claim long texts—it enforces one core per draft and will require pruning down to one first.

Install Agent-Ready

① Verified Mirror 2026-09-15

Option A · Direct DownloadDownload dbs-resonate.tar.gz

sha256: 255f426f94b9e981…

Option B · Copy Guide for Agent

install.json

Agents automatically download and verify mirrors from this JSON guide.

② Upstream GitHub

GitHub Repo ↗

Pinned to commit 8b8e33f1ec.