agentleFS
mrdaioffmrdaioff/baby-brain/.agents/skills/services-catalog/SKILL.md

services-catalog

Skill0 starsChanged 2 months agoMITView on GitHub

Use when creating a new project, ingesting a SOW/proposal/pitch-shaped transcript, brainstorming what to sell a prospect, or when given context about a company and needing ideas for what the owner could offer them. Maintains services/ as a living catalog of concrete, sellable service offerings (not abstract business theory — that stays in knowledge_base/) drawn from real scoped work, signed SOWs, pitches, and brainstorms. Defines the service lifecycle (idea/pitched/scoped/delivered/proven), the catalog schema, and the trigger points where it must be updated automatically.

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 2 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Never mix the two: if you catch yourself writing "the psychology of X" or "why Y works," that belongs in knowledge_base/.

    line 10
  2. Distinctive:

    Categories are yours to invent: Do not copy a list from anywhere — write the five to ten groupings that match what you actually sell, in …

    line 35
  3. Distinctive:

    Never retroactively remove a service that stalled: If a prospect went cold, leave its status where it last had real evidence — don't …

    line 60
  4. Very specific:

    Treating this like knowledge_base: Don't write theory, psychology, or market analysis here — link to the relevant knowledge_base/ entry …

    line 68
  5. Very specific:

    One giant catalog file: Keep one file per service so status/evidence changes are a small diff, same reasoning as one-concept-per-file in …

    line 69

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Services Catalog Protocol

services/ answers one question fast: "Based on everything the owner has actually scoped, pitched, or delivered before, what could we sell this company?" It is the concrete, sellable counterpart to knowledge_base/business/ and knowledge_base/sales/, which capture the abstract theory behind pricing and positioning. A knowledge_base entry explains why a pricing model works; a services/ entry is the menu item itself — name, what it is, what it costs, who's bought it.

Never mix the two: if you catch yourself writing "the psychology of X" or "why Y works," that belongs in knowledge_base/. If you're writing "here is a thing we sell and here's what it costs," that belongs in services/.

Structure

services/
  README.md       — the catalog index: one row per service (Name | Category | Status | Pricing | Proof points)
  <slug>.md        — one file per distinct service

Each services/<slug>.md uses this frontmatter:

---
name: A short, concrete name for the thing you sell
category: <one of your own categories — see below>
status: idea | pitched | scoped | delivered | proven
first_seen: YYYY-MM-DD
last_reinforced: YYYY-MM-DD
origin_projects: [slugs from projects/registry.yml, or "none" if brainstormed only]
related_concepts:
  - "[[knowledge_base/<domain>/<the-theory-behind-it>]]"
---

Categories are yours to invent. Do not copy a list from anywhere — write the five to ten groupings that match what you actually sell, in your own words, and keep them stable once chosen. A category taxonomy borrowed from another business quietly reshapes how you think about your own offering, and it shows up in every proposal you generate from this catalog afterwards.

Body sections: What it is (2-4 sentences) / Pricing model (exact numbers where known, or "not yet priced") / Proof points (which real client engagement used it, with a link to projects/<slug>/ or the source transcript — or "brainstormed only, no client yet") / Related concepts (links into knowledge_base/).

The status lifecycle

Mirrors the knowledge_base emergent → validated → canonical pattern, but scoped to commercial reality:

  • idea — brainstormed (in conversation, a strategy session, or a knowledge_base concept) but never actually proposed to a real prospect
  • pitched — proposed to a specific real prospect, no signed engagement yet
  • scoped — written into an actual SOW or proposal with real pricing attached, for a signed or near-signed deal
  • delivered — actually performed for at least one paying client
  • proven — delivered profitably for 2+ clients; a repeatable, standardized offering

A service only moves forward on real evidence (a SOW, a signed engagement, a second client) — never bump status on optimism. This is the same discipline project-creation uses for active vs prospect.

When to update the catalog (trigger points)

  1. New project created (project-creation skill fires) — read the project's SOW/scope. For each distinct deliverable in it: if it matches an existing services/<slug>.md, bump last_reinforced, add the project to origin_projects, and consider a status bump (e.g. pitched → scoped once a real SOW exists, scoped → delivered once work is actually underway, delivered → proven if this is the 2nd+ client). If it's a genuinely new offering, create a new services/<slug>.md at pitched or scoped depending on whether it's signed.
  2. Ingesting a SOW, proposal, or pitch-shaped transcript (librarian skill fires on ingestion) — same check as above, applied to whatever the document describes, even if no project folder exists yet (e.g. a prospect that never signs still counts as pitched).
  3. A brainstorming or strategy session surfaces a new offer idea (no client attached yet) — add it at idea status so it doesn't get lost, same way knowledge_base captures emergent concepts from a call.
  4. Never retroactively remove a service that stalled. If a prospect went cold, leave its status where it last had real evidence — don't delete, don't silently downgrade.

Using the catalog to brainstorm for a new company

When the user gives context about a company (a new prospect, a call, a research pass) and wants ideas for what to sell them: read services/README.md first and reason from what's already proven/delivered/scoped before inventing something new from scratch. Precedent — "we've sold this exact thing to N companies before" — is a stronger pitch than a fresh idea with zero proof points. Only propose something at idea status if nothing in scoped/delivered/proven fits, and say so explicitly ("this would be new territory — we haven't sold this before").

Anti-patterns

  • Treating this like knowledge_base. Don't write theory, psychology, or market analysis here — link to the relevant knowledge_base/ entry instead.
  • One giant catalog file. Keep one file per service so status/evidence changes are a small diff, same reasoning as one-concept-per-file in knowledge_base/.
  • Forgetting services/README.md. The index table is the fast-scan entry point; a service that only exists as an orphan file is undiscoverable in practice.
  • Bumping status without evidence. A brainstormed idea is not "scoped" until a real SOW says so.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.