Agency

Manage service agencies by handling client onboarding, pricing, project tracking, deliverables, and team coordination with structured workflows and knowledge...

Security Scan

The skill's files and runtime instructions are internally consistent with a local agency operations assistant that reads/writes an ~/agency/ workspace and generates drafts, templates, and reports — it does not request unrelated credentials or install code.

Purpose & Capability

The name/description (agency operations: onboarding, pricing, projects, team coordination) matches the required artifacts and actions: creating client/project files, templates, knowledge base, estimates, proposals, and status reports. There are no unrelated env vars, binaries, or installs requested that would be disproportionate to the stated purpose.

Instruction Scope

The SKILL.md and supporting docs explicitly instruct the agent to create and manage files under ~/agency/ (clients, projects, knowledge, config). They also instruct generating proposals/PDFs and drafting client communications. These actions align with the purpose, but they imply read/write access to the user's home directory and the ability to produce files (including PDFs). The skill explicitly forbids sending proposals or communicating with clients without human approval, which reduces risk. There is no instruction to access other system credentials or external endpoints.

Install Mechanism

Instruction-only skill with no install spec and no code files. That is the lowest-risk install mechanism — nothing is downloaded or written by an installer step.

Credentials

No environment variables, credentials, or config paths are requested. The only resources referenced are local workspace paths under ~/agency/, which are proportional to a tool that manages local agency data.

Persistence & Privilege

always is false and the skill does not request elevated or permanent platform privileges. It prescribes creating and updating files in a user-owned ~/agency/ folder (normal for its purpose). The skill does not request to modify other skills or global agent settings.

Assessment

This skill appears coherent for managing an agency's local workspace, but before installing consider: (1) It will read and write files under ~/agency/ and store client data there — ensure you are comfortable storing sensitive client info in that location and back it up or encrypt if needed. (2) It includes steps to generate proposals and PDFs but provides no tooling; confirm how PDFs will be produced or whether the agent will ask you before using any external service to render or share files. (3) The skill explicitly says not to send proposals or contact clients without human approval; nevertheless, the platform allows autonomous invocation by default — if you want to prevent any autonomous actions, adjust agent/skill invocation controls. (4) If you plan to integrate this with cloud storage, email, or third‑party project tools, require only the minimum credentials needed and review those integrations separately. If you want more assurance, ask the author (or request an updated skill) to declare how documents are rendered/stored and to add an explicit list of any external integrations the skill might use.

When to Use

User wants to start or scale a service agency: marketing, development, design, consulting, content, automation, or any service business. Agent handles operations so human focuses on clients and strategy.

Quick Reference

Area	File
Client onboarding	onboarding.md
Pricing and proposals	pricing.md
Project management	projects.md
Client communication	communication.md
Deliverables workflow	deliverables.md
Team coordination	team.md
Agency-type specifics	by-type.md
Learning system	feedback.md

Workspace Structure

Agency data lives in ~/agency/:

~/agency/
├── clients/           # One file per client
│   ├── index.md       # Client list with status
│   └── [name].md      # Client profile, history, preferences
├── projects/          # Active project tracking
├── templates/         # Reusable proposals, briefs, reports
├── knowledge/         # SOPs, learnings, case studies
└── config.md          # Rates, margins, team structure

Core Operations

Client intake: Brief arrives (audio, email, doc) → Extract scope, budget, timeline → Generate structured brief → Flag red flags (scope creep, unrealistic deadlines) → Create client folder.

Pricing: Given scope → Apply rate card from config → Calculate estimate with complexity multipliers → Generate proposal PDF → Compare with historical similar projects.

Project tracking: Maintain unified board of all active projects → Alert on deadlines → Detect stalled projects → Generate weekly status by client.

Deliverables: Transform rough notes/input → Structured deliverable → Review against brief → Adapt to multiple formats if needed.

Critical Rules

• Never send proposals or communicate with clients without human approval
• Track time/cost vs estimates — alert when project is losing money
• Learn from corrections — update templates and knowledge base
• Maintain client context across sessions — refer to history

Config Fields

Create ~/agency/config.md with rates, team, and margins. See pricing.md for format.