What does this skill do, and when should you use it?
This is the first skill in the emory/ASD-AuDHD-PAI-Skills repository, built for people with PDA (Pathological Demand Avoidance) autism and those who support them. It transforms commands, time pressure, and obligation language into collaborative exploration, real choices, and curiosity-driven phrasing. The skill ships with seven reframing strategies, worked examples across work, healthcare, household, social, and self-care contexts, and a five-step implementation process. It is explicit about authenticity: never fake a choice that isn't real, and the goal is reduced anxiety, not dodged responsibility. The author originally created these skills for their 10-year-old child and notes their own AI assistant, Sundae, has used them to get traction on projects the author had been avoiding.
- Analyzes the core task, time constraints, and real versus perceived urgency in the user's request
- Strips out demand language ("you must/should/right now") and conditional threats
- Selects one or two of seven strategies (choice architecture, collaborative problem-solving, curiosity framing, indirect/third-person, partnership language, time flexibility, autonomy acknowledgment) to produce a reframed version
- Validates the person's autonomy explicitly
- Outputs the rewritten phrasing plus a brief explanation
- Runs no scripts and calls no external tools — it is a pure-instruction SKILL.md
- A parent whose child resists direct instructions for homework or chores, triggering standoff and avoidance
- Someone with PDA stuck for weeks on an internal demand like "I should call the dentist"
- A professional who needs a PDA coworker to meet a deadline without damaging the relationship
- Families seeking lower-anxiety communication around daily care for an autistic child (medication, showering, school mornings)
- Builders of a Personal AI Infrastructure (PAI) who want their assistant to communicate in a demand-avoidance-friendly style
- Someone in crisis or shutdown who needs crisis intervention or safety measures - reframing is a support tool, not a substitute
- Situations requiring accommodations, therapeutic intervention, or executive function support tools (reminders, visual schedules, body doubling) - reframing alone is not sufficient
- Users expecting tested or systematically evaluated outcomes - effectiveness rests on the author's personal experience, and PDA individuals vary widely, so tuning is expected
How do you install this skill?
- Static source review only; nothing was executed. Scores are based on file reading.
- Content touches health-related topics (PDA/autism) without authoritative clinical citations; it should not replace professional medical or therapeutic advice.
- Reframing effectiveness varies by individual — the doc itself admits this — and requires per-person tuning.
- Publisher is unverified; repository has no versioning or changelog, so long-term maintenance is uncertain.
- Core usage depends on the Claude Skills ecosystem; reachability for mainland-China users is not addressed.
Zip the Skills/pda-reframing/ directory, then in Claude Desktop or Claude Mobile go to Settings > Capabilities >> Skills and drag the zip archive into the Skills area. No other installation path is documented.
tmp="$(mktemp -d)"
git clone --depth 1 https://github.com/emory/ASD-AuDHD-PAI-Skills.git "$tmp"
mkdir -p ~/.claude/skills
cp -R "$tmp/Skills/pda-reframing" ~/.claude/skills/
rm -rf "$tmp"Generated from the source repository and skill path; it copies only this skill's folder. If the author's install steps above differ, follow those first. To scope it to one project, replace ~/.claude/skills with that project's .claude/skills.
How do you use this skill?
Once installed, send your agent any of these to trigger it:
- Help me tell my kid to do their homework in a PDA-friendly way
- I have PDA and I've been avoiding 'I should call the dentist' for weeks - reframe it for me
- I need to ask my coworker with PDA to finish an analysis by a deadline - help me phrase it
After installing, trigger it with natural language, e.g. "Help me reframe 'do your homework' in a PDA-friendly way" or "I have PDA — reframe my internal demand 'I should call the dentist'". The skill lists its own trigger conditions: the user mentions PDA or demand avoidance, says "shoulds" and "have tos" make things harder, or is supporting someone with PDA.
What are this skill's strengths and limitations?
- Pure-instruction skill with no scripts or external dependencies — highly portable across clients
- Seven concrete strategies plus rich worked examples spanning work, healthcare, household, social, and self-care contexts
- Explicit ethical guardrails: no manipulation, no fake choices, honest about real constraints and non-negotiable demands
- Acknowledges its limits, listing red-flag situations that require professional support instead of reframing
- Written by a parent of a PDA child, and peer-reviewed plus reviewed by the author's AI assistant before publishing
- No test suite or systematic outcome evaluation; effectiveness rests on the author's personal experience
- The author notes PDA individuals vary widely — the same strategy won't work for everyone and tuning is expected
- Reframing is a support tool, not a substitute for crisis intervention, safety measures, or therapeutic help
- README states other Agent Skills-capable clients (Codex, Aider, gemini cli, etc.) were only "beginning to be supported" at writing time — real-world availability unverified
How does this skill compare with similar options?
Side by side with related skills; every score comes from the same FSRS standard.
| Skill | FS score | Stars | Last updated | License |
|---|---|---|---|---|
| PDA Task Reframing Skill this page | 63 · Recommended | ★ 55 | 10mo ago | MIT |
| ELI5 Audience-Adaptive Explanation Skill | 64 · Recommended | ★ 1.8k | 7mo ago | MIT |
| Kanban AI | 52 · Use with care | ★ 25 | 7mo ago | Apache-2.0 |
| Obsidian CLI Assistant | 46 · Use with care | ★ 50k | 1mo ago | MIT |
| Notion Spec-to-Delivery Planner ✓ OpenAI · Official | 43 · Not recommended | ★ 28k | 4mo ago | — |
How did FollowSkills review this skill?
Pure text prompt skill: no scripts, no external calls, no sensitive-data handling; minimal privilege surface. Explicitly warns against manipulation and demands honesty about real constraints. Deducted for unverified publisher and lack of source attribution/community review evidence.
Instruction-only skill; internal guidance (5-step process, strategies, examples) is self-consistent, no dependencies to break. Deducted to the static ceiling of 10: no tests, no error-handling paths for atypical inputs.
Clear trigger conditions ('When to Use', 'When to Recommend') and explicit non-fit boundaries (crisis, safety, therapeutic needs). Deducted for potentially over-broad semantic triggering (generic procrastination requests) and no discussion of Chinese-language or mainland-China environment fit.
MIT license present, well-layered docs (overview, principles, strategies, examples, ethics, limits), open to contributions. Deducted for no versioning, no changelog, no explicit maintenance responsibility or update path; README is largely anecdotal.
Clear goal (turn commands into low-anxiety phrasing), rich directly-usable examples, clear marginal value. Deducted to static ceiling of 7: output effectiveness unverified and depends on individual variation, which the doc itself acknowledges.
Authorial experiential claims only; no third-party execution or evaluation evidence; PDA-related claims lack authoritative citations and fact/inference separation. Deducted: single evidence type, not independently reproducible.
Open a dimension to read why it scored that way
Evidence confidence：Low — Mostly static review, author material or a limited demo; useful for discovery, not high-risk decisions.
See the full review method →