Tibsfox/gsd-skill-creator / examples/skills/gsd-meta/decision-framework-invoker
decision-framework-invoker
VerifiedTriggers for automatically invoking the decision-framework skill before irreversible or high-blast-radius operations. Use as a checkpoint: when a proposed action matches a trigger pattern, pause and run the appropriate framework (first-principles, inversion, decision matrix, second-order thinking) before proceeding. Prevents ad-hoc calls from becoming regret.
Decision Framework Invoker
The decision-framework skill provides thinking frameworks (First Principles,
5 Whys, Eisenhower, Pareto, Inversion, Second-Order, Decision Matrix). This
invoker skill captures when to use it — the trigger patterns that should
prompt a pause-and-frame before proceeding.
Trigger Patterns
Invoke a decision framework when the proposed action matches any of these:
Irreversibility
- Destructive git operations:
push --force,reset --hard,branch -D - Committing generated content to a shared repo (N files where N > 100)
- Applying a schema migration to a production DB
- Sending a message (Slack, email, PR comment) that cannot be recalled
→ Framework: Inversion ("what's the worst outcome; can we back out?")
Blast Radius
- Changes affecting multiple packages / projects
- Renaming public APIs, CLI flags, config keys
- Removing a feature that has downstream consumers
- Rewriting > 1,000 lines across a codebase
→ Framework: Second-Order Thinking ("what does this cause, and what does that cause?")
Multi-Option Judgment Calls
- Choosing between 3+ architectural paths
- Picking a library/framework with long-tail commitment
- Prioritizing between competing feature requests
→ Framework: Decision Matrix (columns = options, rows = weighted criteria)
Root Cause Debugging
- A bug has returned after being "fixed"
- A pattern keeps recurring across unrelated work
- A test failure has non-obvious cause
→ Framework: 5 Whys (keep asking until you hit the infrastructural cause)
Scoping Ambition
- Feature spec has 20+ items; need to prioritize
- Time-boxed session; need to pick highest-leverage work
→ Framework: Pareto 80/20 ("which 20% delivers 80% of value?")
Reframing Stuck
- Been spinning on an approach that isn't working
- Assumptions feel embedded but might not be true
→ Framework: First Principles (strip to the problem; rebuild from truths)
Invocation Shape
When a trigger fires:
- Name the trigger in plain language: "About to commit 1,881 generated
chapter files — this is the
blast-radius: committing generated contenttrigger." - Pick the framework from the mapping above.
- Apply it in-line in the conversation — usually 3-5 sentences is enough to surface the tradeoff.
- Record the decision via
observe.mjs event decision <label> <payload>so the retrospective captures it. - Proceed with eyes open, or adjust the approach.
Example — The 1,881-File Publish (v1.49 release-history work)
Trigger: "About to commit 1,881 auto-generated markdown files into
docs/release-notes/v*/chapter/."
This hit two trigger categories:
- Blast radius: large diff affecting 1,881 files
- Irreversibility-ish: git commit is reversible, but reverting a commit that size is painful
Framework applied (Inversion):
- Worst case: "These files are regenerable from the DB. If we commit them
and hate the shape, we can
rm -rfand regenerate. Cost: a revert commit." - Best case: "Anyone browsing the repo on GitHub can read chapters without running the pipeline."
- Hidden third option: "Commit only the top-level index files (STORY.md, INDEX.md) and leave per-chapter files local."
Decision: commit all 1,881. Rationale: regenerable, reviewable on GitHub,
user asked. Recorded via observe.mjs event decision.
What This Skill is NOT
- A mandatory gate — you don't need a framework for every tool call
- A substitute for doing the work — frame the decision, then act
- A bureaucratic checklist — most triggers take 30 seconds to resolve
Related
decision-framework— the actual frameworks this skill invokessession-observatory-live— record the decision via observe.mjs
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 468k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 321k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 321k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 321k | Updated | Claude CodeCodex | Verified | Install |