Install
$ agentstack add skill-kumaran-is-claude-code-onboarding-decision-frameworks ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
Iron Law
NO GENERIC ADVICE — every framework must be applied to the specific situation the user describes; "consider your assumptions" is not an application; naming the actual assumptions is
Decision Frameworks
Four single-framework deep-dives for decisions, problems, and commitments. Each mode applies one reasoning framework interactively — asking follow-up questions, pushing for specificity, and refusing vague answers.
Use this skill when you want to go deep on one framework. Use mental-model-applier when you want three frameworks applied in parallel for a broader perspective shift.
When to Use This Skill
- You are stuck on a specific problem and want to reason through it systematically
- A decision keeps feeling hard despite having information — you need a different lens
- You want to challenge your own assumptions before committing
- You are about to make an irreversible choice and want to stress-test it
- You need to make tradeoffs explicit rather than implicit
The Four Modes
| Mode | Framework | Best For | |------|-----------|----------| | First Principles | Break to fundamental truths, rebuild | Overcoming convention, designing from scratch, "why do we do it this way?" | | Inversion | Guarantee failure, then flip | Risk surfacing, strategy validation, pre-commitment checks | | Regret Minimization | Decide from age 80 | Long-horizon personal or career decisions, emotional clarity | | Opportunity Cost | Make tradeoffs visible | Commitments, resource allocation, "should I do this?" decisions |
Mode Selection
Ask the user which framework to apply if they have not specified one. Present the four options with a one-line description. Use AskUserQuestion when interactive mode permits.
If the user's problem maps clearly to one mode, state which you are using and why before beginning.
Core Workflow
- Read the problem — Understand what is being decided or solved. If the description is vague, ask one clarifying question before proceeding.
- Select mode — Match to the problem type using the table above, or ask.
- Load reference — Read the corresponding reference file for the full method and output template.
- Apply interactively — Do not produce a monologue. Ask follow-up questions. Push for specificity. Refuse vague answers.
- Produce structured output — Use the output template from the reference file.
- Offer a second pass — After completing one mode, offer to run a complementary framework.
Reference Guide
| Mode | Reference File | Load When | |------|---------------|-----------| | First Principles | references/first-principles.md | User wants to break assumptions and rebuild | | Inversion | references/inversion.md | User wants to work backwards from failure | | Regret Minimization | references/regret-minimization.md | User faces a long-horizon personal or career decision | | Opportunity Cost | references/opportunity-cost.md | User is evaluating a commitment or resource allocation |
Constraints
MUST DO
- Apply the framework specifically to the user's situation — name their actual assumptions, their actual tradeoffs, their actual failure modes
- Push back on vague answers: "poor execution" is not a failure mode; "the API migration slips 3 weeks because the external team has no SLA" is
- Ask follow-up questions rather than filling gaps with invention
- State which framework you are applying and why before starting
- End with a concrete output the user can act on — not just analysis
MUST NOT DO
- Apply multiple frameworks simultaneously — this skill is for single-framework depth; use
mental-model-applierfor multi-model breadth - Produce generic advice that would apply to any situation
- Accept "it depends" as a final answer — push for the specific version that applies here
- Skip the interactive element and produce a monologue
Complementary Framework Pairings
After completing one mode, suggest a natural follow-on:
| Completed Mode | Natural Follow-On | Why | |----------------|-------------------|-----| | First Principles | Inversion | First Principles finds the right foundation; Inversion stress-tests it | | Inversion | Second-Order Thinker | Inversion finds failure modes; Second-Order maps their downstream consequences | | Regret Minimization | Opportunity Cost | Emotional clarity on direction + explicit tradeoffs of the chosen path | | Opportunity Cost | First Principles | Tradeoffs made visible + assumptions challenged on why those options exist |
Related
> For adversarially stress-testing any decision or plan: use the-fool skill. > For applying three frameworks in parallel for perspective shift: use mental-model-applier skill. > For mapping downstream consequences of a decision: use second-order-thinker skill.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: kumaran-is
- Source: kumaran-is/claude-code-onboarding
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.