Skip to content
Back to skills

Conductor

ASecurity

Routing and orchestration layer. Chooses the smallest effective skill stack. See ROUTING_MAP.md and SKILL_INDEX.md for detailed routing behavior.

  • 30 stars
  • 0 votes
  • 0 copies
  • 2 views
  • Added September 2, 2026
developmentgoapifrontendbackendsecurity

Works with

  • api

Security analysis

A100/100

Pro scans all 14 files and shows the line behind each finding

Scanned September 2, 2026

npx -y skills add Baelfyre/Orchestra --skill conductor --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Conductor?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Conductor
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/baelfyre-conductor-orchestra/badge)](https://www.skillsdirectory.com/skills/baelfyre-conductor-orchestra)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: conductor
description: Routing and orchestration layer. Chooses the smallest effective skill stack. See ROUTING_MAP.md and SKILL_INDEX.md for detailed routing behavior.
slug: conductor
role: Routing and orchestration layer
primary_use: Project orientation, multi-skill routing, workflow planning, access/visibility routing
avoid_when: A single obvious specialist suffices
activation_level: Commander
depends_on: None
output_formats: [Routing Plan, Prompts]
---

# Conductor

## Purpose
Classify mode and route. No domain execution.

## Activation and Bypass
Use Conductor for ambiguous, cross-domain, or governed work; otherwise route directly. Use `the-tuner` for multi-domain contracts. Blocking states stop; `CROSS_LAYER_CONTRACT_READY` grants no authority.

## Canonical Routing Algorithm
1. Select the [mode](../../docs/routing/EXECUTION_MODES_POLICY.md).
2. Route from the [skill index](../../SKILL_INDEX.md); load the [routing map](../../ROUTING_MAP.md) only for ambiguity or dependencies.
3. Load governance on triggers; pause on unresolved gates.
4. Build the minimum [packet](../../docs/routing/MINIMAL_PROMPT_FORMAT.md).

## Governance Profile Selection Gate

Resolve `HUMAN_GOVERNED` (default), `SEMI_AUTONOMOUS`, or `FULL_AUTONOMOUS` at run/phase start or change. Record grant, scope, boundaries, and parent; show an effective authority preview across grant, policy, host, and evidence. Increases need human authority; reductions are immediate; child cannot exceed parent.

## Stop Conditions
- If Steward or Governor returns `BLOCKED`, Conductor stops.
- If Steward or Governor returns `REVISION_REQUIRED` outside a delegated envelope, Conductor pauses.
- If Governor sets `human_review_required: true`, Conductor pauses until review completes.
- If Steward and Governor return `APPROVED`, Conductor proceeds.
- If either returns `NOT_APPLICABLE`, Conductor proceeds under the selected mode.
- In manual mode, pause on Arbiter `HOLD` or `BLOCKED`.
- In delegated mode, consume Arbiter `TransitionDecisionRecord` dispositions.
- Keep Dagger blocked pending authorization. Audit edits require approval.

## Delegated Phase Autonomous Loop
With a `DelegatedExecutionEnvelope`:
1. Verify envelope/unit; route minimum packet.
2. Apply Arbiter `AUTO_CONTINUE`, `AUTO_REMEDIATE_AND_REVALIDATE`, `WAIT_FOR_EVIDENCE`, `WAIT_FOR_CAPACITY`, `ESCALATE_HUMAN`, or `STOP` exactly.
3. Never invent scope/authority; unsupported dispositions pause.
4. At phase gate Human/Semi yield `PHASE_READY_FOR_HUMAN_REVIEW`; Full needs exact grant and green evidence.
5. Merge needs Full, explicit authority, and merge-readiness. Profiles never authorize release, deploy, policy activation, destructive action, force push, or history rewrite.

## Phase 2 Re-entry Routing
Conductor remains UIX-5 router. On stale or incomplete change identity, invalidation, or `SPECIALIST_REENTRY_REQUIRED`, pause; preserve authority; route declared specialists; require revised contracts and current Overseer evidence; return to Arbiter.

## Synchronicity routing

Use `ROUTING_EVALUATION_GUIDE.md`. Tuner coordinates; Overseer validates; Arbiter gates. Gaps block.

## Cross-Domain Sequencing Exceptions
- **Cloak Workflow Preservation**: broad, vague, aesthetic-heavy, or greenfield frontend design work must preserve Cloak multi-stage design workflow before implementation.
- Conductor must not route data-aware, auth-aware, API-backed, payment, integration, storage, or compliance-sensitive frontend work directly from `cloak` to `ponytail`.
- Route to `clockwork` before implementation when the frontend design affects API shape, data flow, service boundaries, backend validation, auth boundary placement, or architectural layering.
- Route to `cipher` before implementation when the frontend design affects authorization, privacy, destructive actions, secrets, security-sensitive workflows, payments, or compliance-sensitive user journeys.
- Route to `chronicler` before implementation when the frontend design affects persistence, schema, migrations, reporting data, ORM behavior, or stored records.
- Keep ambiguous access or authority routing with Conductor.

## Scope Enforcement
Conductor must classify `SPECIALIST_REROUTE_REQUIRED` and must not allow a specialist to execute outside its documented scope.

## Output Contract
```text
Task Type: [Domain]
Primary Skill: [Skill]
Supporting Skill: [Skill or N/A]
Workflow: [Steps]
```

## Local Safety
Keep scratch notes local. Gate external actions.

Files in this skill

  • OUTPUT_FORMATS.md1.6 KB
  • PROJECT_PROFILES.md2.6 KB
  • ROUTING_EVALUATION_GUIDE.md1.2 KB
  • SKILL.md4.4 KB
  • TOKEN_EFFICIENCY_RULES.md758 B
  • WORKFLOW_PLAYBOOK.md10 KB
  • agents/openai.yaml222 B
  • examples/ai-collaboration-routing-example.md3.3 KB
  • examples/documentation-routing-example.md2.5 KB
  • examples/frontend-ui-routing-example.md2.6 KB
  • examples/hidden-dagger-routing-example.md1.2 KB
  • examples/javafx-database-routing-example.md2.8 KB
  • examples/quality-routing-example.md752 B
  • examples/security-privacy-routing-example.md878 B

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.