Skills, knowledge base and tasks for Claude Code, Codex, Cursor, and friends

# Your agent's second brain

How to do everything.

Skills and knowledge, written once as markdown protocols.

How to track what to do.

Tasks your agents and your team file, pick up, and close.

One bootstrap, tuned once, starts every project. Your agent looks up the rest mid-task, then lets it go.

Tell your agent

```
Let's use protocol. Onboard me: https://api.protocol.supply/@protocol/onboarding
```

copy

[Get Started Free](/getting-started)[Browse protocols](/search)

CLAUDE.md · AGENTS.mdbefore · one per repo

```
$ wc -l ~/dev/*/{CLAUDE,AGENTS}.md
   847 ~/dev/api/CLAUDE.md
   412 ~/dev/web/AGENTS.md
   630 ~/dev/billing/CLAUDE.md
   295 ~/dev/infra/CLAUDE.md
  2184 total

$ grep -h "IDs" ~/dev/*/{CLAUDE,AGENTS}.md
- IDs are ashid (repo_, entr_), never UUID
- IDs are UUID v4
- Use ashid for new tables (legacy IDs are UUID)
```

2,184 lines, four copies, already disagreeing

one bootstrapafter · tuned once

```
# every repo's CLAUDE.md or AGENTS.md
For project knowledge, run:
  curly protocol bootstrap

# /settings/bootstrap · tuned once
include:
  @protocol/bootstrap-core
  @you/engineering-defaults
  @you/working-style
  @acme/api-conventions
```

~6K tokens: your defaults + the catalog. The rest loads on demand.

your terminallive · public

```
$ curl -sS https://api.protocol.supply/@protocol/memento # Memento

Run-keyed hand-off notes for Claude Code: write one, clear,
and the fresh session claims it.

Long Claude Code sessions get slow, expensive and worse at
the task, and /compact tends to lose the plot. A memento
fixes that from the inside: before clearing, the session
writes one short file holding exactly what the next session
needs, clears, and reads the file back.

## Do it

auto-memento $CLAUDE_PID <<'EOF'
# <topic>
- **Top of mind** — what you were doing, decided, what bit you
- **Next action** — one concrete thing
- **Owed** — who waits on what
- **Traps** — gotchas that cost you something
- **Pointers** — file:line, short-sha, doc ref. Point, don't pack.
EOF

…
```

No SDK. No MCP server. No 100K-token preload. **Your agent fetches the same way.**

`protocol`

skills + knowledge base

Operational instructions — or a description of anything. Like a skill, but lighter-weight: API-served, searchable, and optionally shared.

`impulse`

tasks

Any task, to-do, rule, mission, or urge. Task management for agents and humans alike — with activity logs, escalation, and repetition.

`rows`

database

A lightweight database on any protocol entry: decisions, lessons learned, records, logs, state. Anything your agents keep, living next to the knowledge.

[

### Screen Time

What's allowed, when, and what we've done about controls. Both kids; adjusted by age.

`https://api.protocol.supply/@any-family/kids/screen-time`Read the protocol →](/@any-family/kids/screen-time)[

### Code Review

Reviews are the highest-leverage hour you spend each week — how the team catches bugs, transfers knowledge, and stays aligned on standards.

`https://api.protocol.supply/@any-corp/engineering/code-review`Read the protocol →](/@any-corp/engineering/code-review)[

### Getting Started

Protocol is on-demand knowledge for AI agents. One-command install, then search and fetch.

`https://api.protocol.supply/@protocol/getting-started`Read the protocol →](/@protocol/getting-started)

`@any-corp`What a 40-person company puts in its brain. Forkable starter pack.

-   handbook/
    -   values.md
    -   decision-records.md
-   engineering/
    -   code-review.md
    -   incident-response.md
-   people/
    -   hiring.md
    -   1-1s.md

[Browse @any-corp →](/@any-corp)

`@protocol`The product's own playbook. How we build agents with protocols.

-   getting-started.md
-   quickstart.md
-   devloop.md
-   model-policy.md
-   walkthrough-build-a-pr-review-agent.md
-   protocol-philosophy.md

[Browse @protocol →](/@protocol)

`@any-family`What a household's shared brain looks like. Forkable, fillable.

-   about.md
-   family/
    -   who-we-are.md
    -   values.md
-   kids/
    -   medical.md
    -   babysitter-brief.md
-   home/
    -   house-rules.md

[Browse @any-family →](/@any-family)

New hire's agent knows your codebase before lunch. Your AI knows your kids' pediatrician.

[Use as a Template](/repos/new)

## MCP servers eat your context window. Protocols don't.

Every MCP server dumps its full tool schema into your system prompt. Five servers = 100K tokens gone before your agent writes a line of code.

### Context Window (200K tokens)

With 5 MCP servers

Overhead: 100K

Work: 100K

With protocol

5K

Work: 195K

"Drowning in PDFs, Slack threads, and meeting transcripts."

Ingest endpoint

Drop in raw → out comes a structured, searchable protocol.

"Skills folder is 47 files, none searchable, nobody knows what's in there."

Skills marketplace

Searchable, versioned, and they only load when an agent fetches them.

"Knowledge dies in someone's Notion. The agent never gets it."

Sync API + import

Mirror your existing docs into a repo in 30 seconds.

"Same answer rewritten across six protocols. Drift."

Include directives

DRY for prompts. <!-- include: @repo/slug --> and you're done.

"No idea what your agent actually read last week."

Per-protocol changelog + read mode

Audit trail for every fetch. Print-ready for the compliance binder.

"Tokens leaking into shell history and agent transcripts."

curly wrapper

Tokens never touch your terminal log.

### Add bootstrap to your existing agent

Drops a one-line directive into your CLAUDE.md, AGENTS.md, .cursorrules, or equivalent so every session starts from your bootstrap.

```
Add protocol to my CLAUDE.md or AGENTS.md: include the line "For project knowledge, run: curly protocol bootstrap".
```

copy

**already onboarded? go further — harvest existing knowledge →**

### Harvest your project knowledge

Pulls your project's code standards and agent harness settings into a fresh protocol repo, organized by topic.

```
Follow the @protocol/harvest-project protocol to harvest my code standards, best practices, and agent harness settings into an organized personal repo on protocol.
```

copy

### Harvest your skills

Converts your skills folder (Claude Code or Codex skills, MCP servers, custom) into protocol entries with summaries, aliases, and proper organization.

```
Follow the @protocol/harvest-skills protocol to convert my skills folder into protocol entries in my personal repo on protocol.
```

copy

### Harvest your workspace

Walks your workspace docs — design docs, RFCs, runbooks, anything markdown — and turns each into a structured protocol.

```
Follow the @protocol/harvest-workspace protocol to harvest my workspace docs into my personal repo on protocol.
```

copy

### Harvest past conversations

Combs your transcripts (Claude Code, Codex, ChatGPT, custom agents) for corrections, lessons learned, preferences, and important details. Files each as a protocol.

```
Follow the @protocol/harvest-transcripts protocol to harvest my past conversations for corrections, lessons learned, preferences, and important details. Organize them into my personal repo on protocol.
```

copy

### It goes to eleven.

You're past the manual. Your team's knowledge lives in protocol, every agent bootstraps it by default, and new hires hit the ground running on day one.

[See impulses in action →](/impulses)

## Free. No credit card. No trial timer.

Every protocol's available. Upgrade when reads pile up.

Lock in current pricing while we're early.

[Get Started Free](/getting-started)[See Pricing](/pricing)