CtrlK
BlogDocsLog inGet started
Tessl Logo

plugin-router

Analyze broad, mixed, or unclear Plugin Factory follow-up requests and select the correct plugin lane. Use when plugin intent lacks a clear lane owner.

55

Quality

62%

Does it follow best practices?

Impact

No eval scenarios have been run

SecuritybySnyk

Passed

No known issues

Optimize this skill with Tessl

npx tessl skill review --optimize ./Plugins/plugin-factory/skills/team_automation/plugin-router/SKILL.md
SKILL.md
Quality
Evals
Security

Plugin Router

Internal helper for Plugin Factory follow-up routing. The canonical front door is [[plugin-factory-router]]; use this skill only after that router selects a router follow-up or when a loaded Plugin Factory workflow explicitly asks for the detailed route map.

Philosophy

  • Route first, execute second.
  • Prefer one clarification over unsafe guessing when ambiguity changes risk.

When to Use

Use when plugin intent remains broad, mixed, or missing a clear lane owner after canonical Plugin Factory routing.

Do not use this as the root Plugin Factory entrypoint. Route root-level plugin lifecycle requests through [[plugin-factory-router]] first.

Inputs

  • request text
  • optional path/source
  • constraints and trust requirements

Outputs

Return a routing handoff object with:

  • schema_version
  • execution_mode
  • selected_lane
  • next_skill
  • required_inputs
  • optional blocked_by
  • confidence

Workflow

Use the detailed routing protocol in references/workflow.md.

Apply the context-disposition policy: move important still-valid context to references, and intentionally discard stale, duplicated, unsafe, superseded, or low-signal text.

Execution Boundaries

Apply the OpenAI-style plugin design contract when ambiguity affects side-effect class, package boundary, child-skill separation, user steering, or plugin output shape. Route once; do not execute lane logic from this helper.

Plugin Router owns lane selection, missing-input reporting, and handoff shape. It does not scaffold, harden, install, refresh projections, mutate marketplace state, or make release-readiness claims.

Read when:

  • You need full lane-selection and handoff protocol details: references/workflow.md.
  • You need side-effect, context-minimization, or user-control routing checks: OpenAI-style plugin design contract.
  • You need to choose whether plugin factory work should build a new artifact, improve an existing one, stay docs-only, or stop: First-principles factory gate.

For non-trivial factory work, include first_principles_gate or an explicit first_principles_gate_status: not_applicable with the reason in the output or handoff before claiming readiness.

Examples

  • "Create a new plugin with marketplace entry." -> route to [[plugin-creator]]
  • "Harden this imported plugin before release." -> route to [[plugin-builder]]
  • "Install this plugin from GitHub and verify visibility." -> route to [[plugin-installer]]

Validation

bash Infrastructure/scripts/validation-and-linting/validate_skill_authoring_family.sh

Fail fast: stop at first failed gate and report blocker text.

Constraints

  • redact secrets and sensitive data by default
  • no lane-specific execution

Anti-Patterns

  • executing lane logic before routing is complete
  • routing by preference instead of explicit intent evidence
  • emitting a handoff without required inputs/missing data

Failure Mode

  • Stop when one request can validly route to multiple lanes with different side-effect classes, ownership boundaries, or validation gates.
  • Ask one blocking question or return blocked_by instead of guessing the lane.

Gotchas

  • Router confidence is not execution authorization.
  • A plugin lifecycle request may contain skill work; route the plugin boundary first, then hand off skill hardening only when ownership is explicit.
  • Marketplace or install intent changes the side-effect class even if the user phrases it as a review.

References

  • references/workflow.md
  • references/contract.yaml
  • references/evals.yaml
  • references/task-profile.json
  • ../../../../../Infrastructure/references/openai-style-plugin-design-contract.md
  • ../../../../../Infrastructure/references/software-literature-expert-lens-pack.md
  • ../../../../../Infrastructure/references/software-literature-skill-expertise-map.md
Repository
jscraik/Agent-Skills
Last updated
Created

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.