architecting-software
Designs software architecture from a confirmed PRD. Use when a PRD exists and architecture must be designed before implementation, when writing ADRs, choosing…
Extracts genuine value from a subject and frames it so the right audience recognises it. Use when producing LinkedIn posts, pitch narratives, cover letters, proposals, case studies, or positioning statements. Not for general writing guidance, negotiation, or copywriting
$ npx -y skills add isvlasov/rageatc-oss --skill persuading --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/persuadingContext preview
The summary Claude sees to decide when to auto-load this skill.
Extracts genuine value from a subject and frames it so the right audience recognises it. Use when producing LinkedIn posts, pitch narratives, cover letters, proposals, case studies, or positioning statements. Not for general writing guidance, negotiation, or copywriting
name: persuading description: Extracts genuine value from a subject and frames it so the right audience recognises it. Use when producing LinkedIn posts, pitch narratives, cover letters, proposals, case studies, or positioning statements. Not for general writing guidance, negotiation, or copywriting templates.
A specialisation of communicating-in-writing for making genuine value visible. Medium freedom: follow **extract → frame → express** in order — skipping extraction and going straight to expression is the most common cause of value communication that fails — but adapt each phase's depth to the mode and context.
Three systematic miscalibrations. They are calibration problems, not character flaws; the methodology corrects them structurally:
Excavate the genuine value before attempting to communicate it.
**Identify the real job being done.** People hire products and people to get a job done, not for features and credentials:
Most value communication addresses only the functional job; the emotional and social jobs are often stronger motivators. Surface all three and identify which is primary for this audience.
**Map the competitive landscape** — Dunford's five questions:
1. What would the audience do if this didn't exist? (real alternatives, not imagined ones) 2. What does this offer that those alternatives genuinely lack? 3. What outcomes does that translate to — not capabilities? 4. Which audience segment genuinely cares about that value? 5. Which framing makes the strengths self-evident before any argument is made?
The alternatives question matters most: people never evaluate anything in isolation. Control what they compare against, and you control the evaluation.
**Apply the "So What?" test to every claimed benefit** until you reach a genuine human outcome: "Five years in distributed systems" → so what? → "Deployment failures down 60% in three production environments" → so what? → "The team ships faster and the on-call rotation sleeps." The end of that chain is what belongs in the communication.
**Clear curse-of-knowledge symptoms:** any term the audience doesn't know, any context they don't have — what would a stranger reading this cold be confused by?
Determine the evaluative context before choosing what to say.
**Choose the right category.** Category choice triggers assumptions — about price, alternatives, expectations — before any other communication occurs. "Senior developer" activates different assumptions than "platform engineer who has led three zero-downtime migrations — the person you call when the deploy can't fail": the second narrows the category, names the outcome, and implies the audience. Be the best option in a narrower frame rather than an average option in a broad one.
**Control the reference point.** Decide explicitly what the audience should compare this against: the status quo as benchmark, then the improvement — or the cost of inaction, then the alternative. Loss framing often motivates more than gain framing, but the right choice depends on the audience's current state. The reference point is a choice, not a given.
**Position the subject as guide, not protagonist.** The audience is the hero. Instead of "I have 10 years of experience in X", name the audience's problem and the transformation they want, show you understand it, then provide the evidence that makes you the credible guide to it.
**Sequence emotion before evidence.** The audience's felt response shapes their reception of the argument before they consciously consider it. Order: (1) show you understand their situation, (2) surface the problem or tension, (3) deliver the evidence and argument. Evidence before emotional recognition lands poorly — the audience hasn't yet agreed there is a problem worth solving.
**Apply the SUCCESs diagnostic** before and after drafting:
| Dimension | Test | Common failure | |-----------|------|----------------| | Simple | One core claim — the single thing the audience must remember | Too many messages | | Unexpected | Does the opening violate a predictable expectation? | Predictable framing | | Concrete | Specific numbers, names, outcomes, sensory details? | Abstract language | | Credible | Can the claims be verified? Is there proof? | Unverifiable assertions | | Emotional | Does something connect to what the audience cares about? | Rational-only appeals | | Stories | A narrative arc — situation, complication, resolution? | Lists of facts |
Mediocre content actively harms credibility in saturated contexts; rare excellent content compounds. T
Rage Against The C - pick your own C to rage against. Two plugins for Claude Code / Cowork, built on the idea that we're using AI wrong: the speed of its output tricks us into rushing the input.
Repo: isvlasov/rageatc-oss
Designs software architecture from a confirmed PRD. Use when a PRD exists and architecture must be designed before implementation, when writing ADRs, choosing…
Writes correct, version-aware Telegram bot code. Use when writing, extending, or debugging a Telegram bot in python-telegram-bot, aiogram, grammY, or Telegraf.…
Converts an approved ARCHITECTURE.md into an implementation roadmap of isolated, dependency-ordered chunks. Use when architecture has been approved and work…
Delegates a task to OpenAI Codex running as an interactive session in a herdr pane - uses the user's ChatGPT subscription, visible in herdr, steerable…
Delegates a task to a local LLM running as a Pi coding-agent session in a herdr pane - the subagent is visible in herdr, can be steered mid-session, and costs…
Creates a design system for software with a UI. Use when a project has a user interface and architecture is confirmed — whether creating from scratch or…