Workflow & Core Tools · dontbesilent2025/dbskill

dbs-action

Name: dbs-action
Author: dontbesilent2025/dbskill

Diagnoses execution paralysis using an Adlerian psychology framework. Used when users know what to do but repeatedly delay, procrastinate, or abandon actions.

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

Author dontbesilent2025GitHub dontbesilent2025/dbskill ↗Stars 10121License CC BY-NC 4.0（仓库根 LICENSE 首行「Attribution-NonCommercial 4.0 International (CC BY-NC 4.0)」，条款为 Attribution + NonCommercial，Copyright (c) 2026 dontbesilent。**非商业授权**：个人/学习/研究可直接使用，商业用途需单独授权；GitHub API license.spdx_id 为 NOASSERTION（CC 系非 SPDX 标准标识））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

Pure prompt diagnostic engine: Rooted in 5 explicit axioms (Adlerian Individual Psychology + author's field observations), it first determines the root cause of 'why no action is taken' via axioms before drilling down into concrete signals.

skills/dbs-action/SKILL.md

「想赚钱」的意思就是不想赚钱，「正在赚钱」的意思才是想赚钱。用行动定义意愿，不要用语言定义意愿。

What happens here: Procrastination is redefined from an ability issue into a purposive issue—inaction serves to protect the persona of 'I am capable, just hindered by procrastination' (Axiom 1). Hence, the diagnosis does not ask 'how to start', but 'what consequence are you avoiding?'. The 5 axioms serve as common premises for all reasoning, and every subsequent signal diagnosis traces back to one of them.

Phase 1 enforces listening before diagnosing: Asks only one concrete, clarifying question and then stays silent, preventing premature conclusions when information is insufficient.

skills/dbs-action/SKILL.md

问用户：**「你现在卡在什么地方？说具体的。」**

The template explicitly instructs: 'Then shut up and listen. Do not rush to diagnose. Let the user finish.' This is an anti-jumping-to-conclusions constraint—while most 'diagnostic' prompts dispense advice while asking questions, this separates questioning and judgment into distinct stages.

Phase 2 uses a six-signal pattern matrix (A Execution Simulator / B Intellectual Masturbation / C Direction Hopping / D Knowledge Addiction / E Perfectionism / F Coerced Narrative), each structured in a rigid three-part format: behavioral checklist → diagnosis → one-line prescription.

skills/dbs-action/SKILL.md

**诊断**：模拟到执行的比例是 1/2000。

Each section lists verbatim user examples (e.g., 'Let me think about it more', 'I'll start when I'm ready', 'This doesn't fit me, let me try another'), transforming 'identification' into pattern matching against concrete phrasing rather than vague impressions. The fixed three-part structure ensures consistent, comparable outputs.

Phase 3 converges into a fixed report template: Problem Restatement → Triggered Signal → Diagnosis → The Real Problem → Adlerian Solution → One-line Prescription → Disclaimer.

skills/dbs-action/SKILL.md

这是一个基于 dontbesilent 推文逻辑的 AI 诊断工具，不是心理咨询。

The template is embedded inside code blocks in SKILL.md (# Execution Diagnostic Report / ## The Problem You Described … ## ⚠️ Disclaimer). This enforces output discipline: delivering memorable one-line takeaways like 'You are not preparing to execute; you are substituting preparation for execution' as shareable, screenshot-ready deliverables.

The solution relies on an Adlerian 'helping others' action prescription rather than pep talks, providing a 4-step actionable sequence to derive momentum from social contribution.

skills/dbs-action/SKILL.md

帮助他人。当你帮助别人解决问题时，你会获得社会认同感，你会发现自己是有价值的。

The 4 steps are: Find someone who needs help more than you → Use existing knowledge to solve one concrete problem → Gain proof of 'I am valuable' from positive feedback → Use this momentum to kickstart your own action. Unlike generic encouragement, it offers concrete steps that can be carried out on the same day.

Anti-chicken-soup / anti-rationalization prohibition list: Explicitly bans comforting platitudes, attributing inaction to insufficient information, or pretending to replace a professional therapist.

skills/dbs-action/SKILL.md

不要假装你能替代心理咨询师——标注免责声明

There are 5 strict prohibitions (Do not say you're already doing great / Do not say everyone has their own pace / Do not help the user search for more prep work / Do not attribute inaction to lack of info / Do not pretend to replace a therapist). This hardens the 'tone' into auditable rules, preventing the model from slipping into generic motivational rhetoric.

2Core Capabilities

01Identification and categorization of six execution paralysis signals (Simulator / Overthinking / Direction Hopping / Knowledge Addiction / Perfectionism / External Attribution)

02Translating 'inaction' into 'what is being protected' using teleology, outputting a singular judgment on 'what the real issue is'

03Structured diagnostic report with a one-line prescription, ready for direct archiving and screenshotting

04Adlerian action prescription: 4-step sequence ('Help others → Gain social validation → Kickstart yourself') instead of comforting platitudes

05Anti-chicken-soup guardrails: Bans comforting rhetoric, forbids rationalizing procrastination, and enforces a psychological counseling disclaimer

06Inline positive and negative case libraries (each with 'diagnostic key points'), grounding abstract axioms in concrete scenarios

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; no script execution or external impact.

Audit Notes: The entire directory contains no scripts, network calls, credential access, or local write operations. SKILL.md explicitly declares: 'This is not an encouragement tool. This is a diagnostic tool.' All actions consist of conversational diagnostic language executed via the host model; this skill neither declares nor implements any tool permissions. The only consideration is on the content side: it uses Adlerian 'teleology' to attribute procrastination directly to inferiority complexes or intentional avoidance while strictly prohibiting comforting platitudes, which may be blunt for emotionally fragile users. However, SKILL.md includes a disclaimer and referral guidance for professional therapy, which relates purely to stylistic persona and presents zero code-level security risks.

5Second-Pass Verification

[ok] 无脚本/无网络/无凭证/无写入（green 档依据） — 目录内仅 SKILL.md 与 agents/openai.yaml；token 扫描 http|curl|wget|API_KEY|token|secret|\.env|bash|exec|python|~/. 全部零命中；SKILL.md 无任何读写文件或调用工具的指令，唯一外部动作是结尾提示用户可输入 /dbs。
[ok] 「六信号」确实为 6 类且各有诊断 — 信号 A 执行模拟器 / B 思考手淫 / C 方向跳跃 / D 知识上瘾 / E 完美主义 / F 被强迫叙事，逐一读过，每类含表现清单、诊断、一句话三段。
[ok] 免责声明存在且位置正确（报告模板末段） — 报告模板含「## ⚠️ 免责声明」，正文另有一条禁令「不要假装你能替代心理咨询师——标注免责声明」。两条互相对应。
[ok] 是否夸大（自称「诊断工具」是否只是安慰话术的包装） — 实地核对：报告模板强制输出「真正的问题是什么」与「一句话处方」，风格禁令禁止安慰语；机制与声明一致。但需注意它并不做任何事实查证或心理测评，结论强度靠断言而非证据——已在 risks 备注。
[ok] 与 marketplace 条目描述一致 — .claude-plugin/marketplace.json 中 dbs-action 描述为「dontbesilent 执行力诊断。用阿德勒心理学框架诊断你「知道该做什么但就是不做」的真正原因。」与 SKILL.md description 语义一致，无额外承诺。
[ok] 元数据（license/stars/last_push/installs/commit） — GitHub API：stars 10121、pushed_at 2026-09-07T03:48:59Z、license.spdx_id NOASSERTION（仓库根 LICENSE 为 CC BY-NC 4.0，非商业）；本地 HEAD = 8b8e33f1ecaed8cee606fe950c4426b525ead314 = 任务 pin。
[discrepancy] meta.installs 值域校正（批4 质检 2026-09-15） — 原记录为 skills.sh 仓库级聚合徽标（579.6K（skills.sh 徽标 /b/dontbesilent2025/…），并非本 skill 的单体安装量；站内惯例只记可定位的单 skill 数值 → 校正为 null（其他 25 份 dbs 报告本就为 null，修此 8 份以齐平）

6Verdict & Scenarios

Refuses comfort, provides only diagnosis: Deconstructs 'why you don't act' into named signals and axioms, delivering analytical judgment rather than emotional massage.

Identifies patterns based on verbatim user quotes: Supplies typical phrasing examples for each signal ('Let me think more', 'I'll start when ready'), reducing arbitrary labeling by the model.

Actionable same-day solutions: Proposes a concrete 4-step sequence to help others rather than abstract mindset shifts.

Fixed output structure: Embedded report template ensures conclusions can be easily captured in screenshots and tracked over time.

Zero-code, zero-asset: Contains no scripts, network calls, or write operations; portable across any host environment.

Best for: Ideal for diagnosing execution paralysis when someone 'knows what to do but cannot move forward', especially for entrepreneurs or content creators stuck in direction hopping, over-preparation, or knowledge hoarding: obtain a classification, a one-line prescription, and a same-day action item in a single dialogue. Also serves as a methodology style sample for skill authors wishing to understand the dbskill style (strong assertions + fixed report template + case studies). Best experienced as a quick, single-session self-diagnosis when encountering clear execution bottlenecks.

Not for: Not suitable as a replacement for psychological counseling or therapy (SKILL.md explicitly states 'This is not therapy'); not suitable for situations requiring gentle emotional support or passive listening rather than judgment; not suitable for users seeking factual root-cause analysis (it conducts no investigation and does not differentiate non-psychological factors like ADHD, clinical depression, or resource limitations); not suitable for team managers diagnosing why an employee failed to execute, as it is strictly designed for first-person self-reflection.

Install Agent-Ready

① Verified Mirror 2026-09-15

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

sha256: 2cc9c2f3f243ba21…

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.