- Home
- Skills
- Outfitter Dev
- Agents
- Pathfinding
pathfinding_skill
- TypeScript
25
GitHub Stars
1
Bundled Files
2 months ago
Catalog Refreshed
4 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill outfitter-dev/agents --skill pathfinding- SKILL.md8.0 KB
Overview
This skill helps turn ambiguous or incomplete requirements into a clear, actionable path. It guides discovery, frames focused questions, and produces staged outputs from Prep through Deliver so teams can move forward with confidence. Use it to brainstorm, clarify scope, and decide next steps when the goal or constraints are uncertain.
How this skill works
The skill calibrates an initial confidence level based on provided context, then creates a stage (Prep, Explore, Clarify, or Deliver) and advances stages only forward. It asks focused, structured questions, proposes 2–3 approaches with trade-offs, and recommends a path. At level 5 it delivers an artifact; below 5 it lists caveats and next clarifying questions.
When to use it
- Requirements are ambiguous or incomplete and you need to define the path forward
- Exploring multiple design/implementation options for a complex feature
- Starting a greenfield project where scope and constraints are unknown
- Collaborative brainstorming or problem-solving sessions with engineers/product
- When you want staged progress tracking and explicit confidence levels
Best practices
- Provide as much context up front (docs, previous convo, constraints) so confidence starts higher
- Answer clarifying questions one at a time; the flow expects a single focused question per turn
- Prefer the recommended ★ option but feel free to combine or tweak choices
- Treat confidence levels honestly—don’t force delivery from level 0–3
- Use the skeptic check at high-risk moments to catch hidden assumptions
Example use cases
- You have a vague feature request: clarify scope, list unknowns, and produce an implementation outline
- Comparing two backend approaches: get pros/cons, a recommendation, and next validation steps
- Greenfield API design: identify constraints, success criteria, and a phased delivery plan
- Team planning: surface risks, create stage-based tasks, and produce a delivery artifact
- When a stakeholder asks to “figure out” feasibility before estimating work
FAQ
Confidence sets the starting stage and next actions: low confidence triggers clarifying questions; high confidence can skip to Clarify or Deliver with caveats.
What happens if significant risks appear mid-session?
The skill surfaces the concern with △, asks how to proceed, and may escalate via pushback (◇/◆/◆◆) or invoke the skeptic agent.