zotfiles.md Raw History Folder

zotfile agents

A zotfile packages an agent's behavior into one portable .zot file. It can contain the agent's instructions, reusable skills, static assets, and metadata describing the runtime, model, operating-system, binary, and tool permissions it needs.

The current implementation supports creating, inspecting, verifying, and running local directories and .zot archives. It can also run an agent repository or directory directly from any public GitHub repository without keeping a clone. zot has no built-in owner or official collection. Indexed registry distribution, installation, signatures, bundled executable extensions, network permissions, and environment permissions are not implemented yet.

Quick start

Create a directory for the agent:

reviewer/
├── manifest.json
├── AGENT.md
└── skills/
    └── code-review/SKILL.md

Add a minimal manifest.json:

{
  "zotfile": 1,
  "name": "code-reviewer",
  "version": "0.1.0",
  "description": "Reviews a repository and reports actionable findings.",
  "runtime": {
    "min_zot": "0.2.76"
  },
  "model": {
    "requires": ["tools"],
    "min_context": 64000
  },
  "permissions": {
    "fs": {
      "read": ["${workspace}"],
      "write": []
    },
    "bash": {
      "mode": "none"
    }
  },
  "requirements": {
    "os": ["darwin", "linux", "windows"]
  },
  "entry": {
    "greeting": "What should I review?",
    "default_prompt": null
  }
}

Add the agent's standing instructions to AGENT.md:

# Code reviewer

Review the current repository without modifying it.

Prioritize correctness, security, regressions, and missing tests. Report findings
in severity order with file and line references. Do not praise the code or
summarize files that have no findings.

Test the directory directly during development:

zot inspect ./reviewer
zot run ./reviewer
zot run ./reviewer "Review the authentication package"

Package it when ready:

zot pack ./reviewer
# wrote code-reviewer.zot
# digest sha256:...

zot verify ./code-reviewer.zot
zot run ./code-reviewer.zot

The first run displays the declared permissions and asks for consent before the agent starts.

Directory layout

my-agent/
├── manifest.json          # identity, requirements, and permissions, required
├── AGENT.md               # persona and standing instructions
├── skills/                # SKILL.md directories, optional
│   └── my-skill/SKILL.md
├── assets/                # static files, optional
└── README.md              # human-facing documentation, optional

Only manifest.json. Other ordinary files and directories are included in the archive, but the runtime gives no special behavior to assets/ or README.md yet. Files in assets/ are not automatically added to the model context or granted filesystem access.

Do not include executable extensions in extensions/. The local runtime rejects any bundled extension directory containing extension.json because extension subprocesses cannot yet be confined to the manifest permissions. This is a deliberate fail-closed restriction.

Symlinks are not supported and cause zot pack to fail.

`AGENT.md`

AGENT.md is optional. When present, it is appended to zot's system prompt by default. It should describe the agent's role, workflow, constraints, output format, and when to load bundled skills. When absent, the system prompt is left unchanged.

Keep capability and security declarations out of this file. Permissions come only from manifest.json and are enforced independently of agent-authored prose.

Set replace_system_prompt to true in the manifest to use AGENT.md as the replacement system prompt:

{
  "replace_system_prompt": true
}

Replacement is intended for fully specialized agents. The default layering behavior is usually preferable because it retains zot's normal identity and tool-use guidance. With replacement enabled, an empty AGENT.md intentionally omits the built-in identity while retaining generated context.

Bundled skills

Place skills under skills/<name>/SKILL.md using the normal zot skill format:

skills/
└── investigate-failure/
    └── SKILL.md

---
name: investigate-failure
description: Diagnose a failing command from its output and relevant source files.
---

# Investigate failure

1. Reproduce the failure if permissions allow it.
2. Identify the first meaningful error rather than downstream noise.
3. Read the implementation and nearest tests.
4. Propose the smallest correction and validation plan.

Bundled skills are added to normal skill discovery while the zotfile is running. The model sees their name and description in the skill manifest and can load the full body through the skill tool. See skills.md for the complete skill format.

Manifest reference

The current manifest shape is:

{
  "zotfile": 1,
  "name": "code-reviewer",
  "version": "0.1.0",
  "description": "Reviews a repository and reports actionable findings.",
  "license": "MIT",
  "runtime": {
    "min_zot": "0.2.76"
  },
  "model": {
    "requires": ["tools", "reasoning"],
    "min_context": 64000,
    "preferred": ["claude-opus-4-5", "gpt-5.5"],
    "min_tier": ""
  },
  "permissions": {
    "fs": {
      "read": ["${workspace}", "${agent_data}"],
      "write": ["${agent_data}"]
    },
    "bash": {
      "mode": "allowlist",
      "allow": ["git", "go"]
    },
    "net": {
      "allow": []
    },
    "env": {
      "read": []
    }
  },
  "requirements": {
    "bin": ["git", "go"],
    "os": ["darwin", "linux"]
  },
  "entry": {
    "greeting": "What should I review?",
    "default_prompt": null
  },
  "replace_system_prompt": false
}

Top-level fields

Field	Required	Current behavior
`zotfile`	yes	Format version. Must be `1`.
`name`	yes	Local agent identity and session/data namespace. See naming rules below.
`version`	no	Displayed in consent and inspection output. Semver is recommended but not currently validated.
`description`	no	Displayed by `zot inspect`.
`license`	no	Package metadata. Not otherwise interpreted by the local runtime.
`runtime.min_zot`	no	Minimum zot version. Older binaries refuse to run the agent. Unversioned development builds cannot satisfy a non-empty minimum.
`model`	no	Model capabilities, minimum context, and preferences.
`permissions`	no	Filesystem and bash permission ceiling. Omitted scopes deny access.
`requirements`	no	Required operating systems and executables.
`entry`	no	Initial presentation and prompt metadata.
`replace_system_prompt`	no	Replaces rather than appends to the normal system prompt when `true`. Defaults to `false`.

Name rules

The local runtime currently accepts a flat lowercase name containing only:

a to z
0 to 9
., -, and _

Examples:

code-reviewer
repo_onboarder
acme.audit

Names containing /, uppercase letters, spaces, or path-like values such as .. are rejected. Registry-style namespace/agent names are reserved for the future distribution runtime and are not accepted locally yet.

Model requirements

model.requires currently recognizes:

Capability	Behavior
`tools`	Accepted. Every model in the current catalog is treated as tool-capable.
`reasoning`	Requires a catalog model marked as supporting reasoning.
`vision`	Recognized, but currently fails closed because vision support is not represented in the model catalog. No model qualifies.

Unknown capabilities are rejected. Use tools, not the draft spelling tool_calling.

model.min_context requires a catalog model with at least that many context tokens.

model.preferred is an ordered list of model IDs. It is a preference, not a provider requirement. Selection proceeds as follows:

An explicitly selected model, if present, must satisfy the requirements.
The user's configured default is retained if it satisfies them.
The first compatible model in preferred is selected.
Otherwise, zot chooses a compatible active catalog model.
If none qualifies, execution stops with an error.

model.min_tier is present in the format shape but is not supported by the local runtime. Leave it empty or omit it. A non-empty value is rejected.

Runtime and host requirements

Use requirements.os to restrict the agent to Go operating-system names:

{
  "requirements": {
    "os": ["darwin", "linux", "windows"]
  }
}

Use requirements.bin for commands that must already be on PATH:

{
  "requirements": {
    "bin": ["git", "python3"]
  }
}

Zot checks these requirements before requesting consent. These zotfiles do not have install or postinstall hooks, so authors must document how users can obtain missing programs.

Entry fields

entry.pre is auto-submitted once when zot run starts, before the initial prompt is applied. Values that begin with ! use the interactive shell-escape path (or BashTool in --print / --stream / --json). Shell output streams live to the TUI (interactive) or stderr (non-interactive). Other values are sent as a normal user turn; in --stream mode that turn's assistant text also streams to stdout. After pre finishes, zot reloads extensions and rediscovers skills so anything installed by that command is available for the following turn.

{
  "entry": {
    "pre": "!npx -y skills add tt-a1i/archify --skill archify --agent cursor --copy --yes"
  }
}

entry.default_prompt is used when the user runs the agent without a prompt. It may be a string or null:

{
  "entry": {
    "default_prompt": "Review the current repository"
  }
}

In interactive mode, when no prompt is given on the command line, default_prompt pre-fills the editor after pre finishes (or immediately when pre is empty). It is not auto-submitted. A prompt supplied on the command line takes precedence the same way.

In --print / --stream / --json modes, pre still runs first: shell escapes abort the run on failure; plain-text pre is sent as a first agent turn before the main prompt (default_prompt or the CLI prompt).

entry.greeting is part of the manifest, but the current local runtime does not render it yet.

Filesystem permissions

Filesystem permissions are enforced by the built-in read, write, and edit tools. Empty or omitted scopes deny the operation.

Two variables are available in filesystem scopes:

Variable	Resolves to
`${workspace}`	The current working directory used for the run.
`${agent_data}`	The agent's persistent private data directory under `$ZOT_HOME/agents/<name>/data/`.

Relative paths are resolved beneath ${workspace}. For example:

{
  "permissions": {
    "fs": {
      "read": ["src", "go.mod"],
      "write": ["${agent_data}"]
    }
  }
}

This allows reads under <workspace>/src and at <workspace>/go.mod, plus writes under the private agent data directory. It does not grant general workspace writes.

Common permission profiles follow.

Read-only repository agent:

{
  "permissions": {
    "fs": {
      "read": ["${workspace}"],
      "write": []
    },
    "bash": { "mode": "none" }
  }
}

Repository editor with private persistent state:

{
  "permissions": {
    "fs": {
      "read": ["${workspace}", "${agent_data}"],
      "write": ["${workspace}", "${agent_data}"]
    },
    "bash": { "mode": "none" }
  }
}

Data-only agent:

{
  "permissions": {
    "fs": {
      "read": ["${agent_data}"],
      "write": ["${agent_data}"]
    },
    "bash": { "mode": "none" }
  }
}

Filesystem checks canonicalize existing paths, or the nearest existing parent for a new path, so symlinks cannot be used to escape a declared scope.

If the working directory changes during a run, ${workspace} permissions are expanded again for the new directory.

Bash permissions

permissions.bash.mode supports three values.

`none`

{
  "bash": { "mode": "none" }
}

All bash tool calls are denied. This is also the default when mode is omitted.

`ask`

{
  "bash": { "mode": "ask" }
}

Bash is available after the user approves the agent at launch. Consent for an agent using ask is requested on every launch and is not cached.

In the current local runtime, ask is launch-time capability consent. It does not open a separate confirmation dialog for every individual bash invocation.

The optional Windows powershell tool also requires bash.mode: "ask" for packaged agents. It is denied in all other permission modes and while jailed. Bash allowlists cannot safely validate PowerShell syntax. Enable the tool via the Windows-only PowerShell tool toggle in /settings or select it explicitly with --tools read,write,edit,glob,powershell; see PowerShell tool.

`allowlist`

{
  "bash": {
    "mode": "allowlist",
    "allow": ["git", "go", "grep"]
  }
}

Every command in a shell expression must be listed. Zot checks commands separated by ;, &&, ||, and pipes. Paths are reduced to their base command name, so /usr/bin/git is checked as git. A leading environment assignment is skipped when identifying the command.

The allowlist rejects shell substitution, newlines, redirection, backticks, $(), <, and >. This intentionally supports simple command pipelines rather than arbitrary shell programs.

An allowlist mode with an empty allow array is invalid.

Only declare commands the agent genuinely needs. requirements.bin checks that a command exists, while permissions.bash.allow controls whether the agent may invoke it. Agents that execute a required command normally need it in both places.

Network and environment permissions

The manifest shape includes permissions.net.allow and permissions.env.read, but the local runtime does not enforce them yet. To avoid presenting unenforced declarations as security controls, it rejects manifests containing any network host or environment variable.

Use empty arrays or omit these sections:

{
  "permissions": {
    "net": { "allow": [] },
    "env": { "read": [] }
  }
}

Bundled executable extensions are rejected for the same reason. Network filtering, scrubbed extension environments, and extension confinement belong to a later runtime milestone.

Before the first run, zot prints the agent identity and expanded filesystem and bash permissions, then asks:

Agent code-reviewer@0.1.0 wants to run.

  fs read: /path/to/workspace
  fs write: none
  bash: none

Allow? [y/N]

For modes other than bash: ask, approval is cached for the exact artifact digest under:

$ZOT_HOME/agents/<name>/consents/<digest>.json

Any change to the packaged artifact produces a different digest and requires consent again. bash: ask always requires fresh launch consent.

Non-interactive runs refuse to bypass consent by default. For controlled automation, pass -y / --yes or set:

zot run -y ./code-reviewer.zot --print "Review this repository"

ZOT_AGENT_CONSENT=1 zot run ./code-reviewer.zot --print "Review this repository"

This environment variable skips the consent prompt. Only use it after independently inspecting and trusting the exact artifact being run.

Persistent agent data lives under:

$ZOT_HOME/agents/<name>/data/

The directory is created for every run, but the agent can access it only when ${agent_data} appears in the relevant filesystem permission scope.

Agent-scoped sessions

Sessions created by a zotfile are isolated from ordinary zot sessions and from other agents:

$ZOT_HOME/sessions/agents/<name>/

Normal session flags still apply, including --continue, --resume, --session, and --no-session, but their default storage root is scoped to the active agent. The interactive /sessions picker and /session import, fork, and tree commands use that same agent-scoped root.

Commands

`zot run`

Run an unpackaged directory during development:

zot run ./my-agent
zot run -y ./my-agent
zot run ./my-agent "Do the task"
zot run ./my-agent -y --print "Do the task"

Run a packed artifact:

zot run ./my-agent.zot
zot run ./my-agent.zot --print "Do the task"
zot run ./my-agent.zot --stream "Do the task"
zot run ./my-agent.zot --json "Do the task"

Arguments after the reference use zot's normal CLI parser, so model, provider, reasoning, cwd, session, tool, and output-mode flags remain available. The manifest still imposes its permission and compatibility ceiling. In interactive mode, enabling show loaded resources at startup in /settings displays the manifest name in an [Agent] section above the transcript.

Local filesystem directories, local archive paths, GitHub shorthand, and public GitHub agent-directory URLs are accepted. A single-part name resolves only to a matching local directory or .zot archive:

zot run my-agent

A two-part name is treated as a GitHub owner/repository reference. Resolution remains local-first, then uses the repository root:

zot run frkr/zot-archify

Use three or more parts to select an agent directory from any public GitHub collection:

zot run acme/agents/reviewer
zot run acme/clankers/reviewers/go

There is no built-in owner, official collection, collection configuration, or allowlist. Every shorthand part must use lowercase letters, digits, dots, hyphens, or underscores. Backslashes are also accepted as separators on Windows. Prefix a reference with ./, use an absolute path, retain the .zot suffix, or provide a complete GitHub URL to force local or explicit remote resolution.

For GitHub, zot downloads the repository archive into a temporary directory, selects the requested agent subdirectory, validates it, runs it, and removes the downloaded files when the command exits:

zot run https://github.com/acme/agents/reviewer

A standard GitHub tree URL is also supported:

zot run https://github.com/acme/agents/tree/main/reviewer

Short forms read the repository's default branch through GitHub's HEAD archive. A tree URL uses the branch or tag in the URL. Private repositories and branch or tag names containing / are not currently supported. The downloaded source is temporary, but normal agent data, consent receipts, and session transcripts remain under $ZOT_HOME.

owner/repository and owner/repository/path are direct GitHub mappings, not an indexed or signed registry. Installed names, arbitrary URLs, and OCI references are not resolved yet.

`zot pack`

zot pack [directory] [output]

Examples:

zot pack
zot pack ./my-agent
zot pack ./my-agent ./dist/reviewer.zot

The default directory is the current directory. When output is omitted, zot uses <manifest-name>.zot. If the output has no .zot suffix, zot adds one.

Packing validates the manifest and directory, then creates a zstd-compressed tar archive. Entries are sorted and tar metadata is normalized with fixed timestamps and numeric ownership. The output archive itself is excluded when it is located inside the source directory. Zot prints the SHA-256 digest of the resulting compressed artifact.

`zot inspect`

zot inspect ./my-agent
zot inspect ./my-agent.zot
zot inspect https://github.com/acme/agents/code-reviewer

Inspection validates and prints the agent's name, version, description, digest, declared permissions, and complete file list. It does not execute the agent.

For a directory, the digest is computed from its canonical uncompressed tar representation. For a .zot archive, the digest is computed from the archive bytes. Therefore a source-directory digest and its packed-archive digest are not expected to match.

`zot verify`

zot verify ./my-agent.zot

The local verify command validates that the archive can be safely loaded, validates the manifest and required layout, and prints its SHA-256 digest.

Despite the command name, it does not verify a publisher signature in the current runtime. Signature and namespace verification are planned for registry distribution.

Packaging and extraction limits

The archive loader applies these limits:

Limit	Value
Compressed `.zot` file	100 MiB
One extracted entry	64 MiB
Total extracted content	256 MiB

Archive extraction rejects absolute paths and parent traversal. Unsupported tar entry types are ignored. New packages use zstd; gzip-compressed tar archives are accepted only as a compatibility fallback for earlier experiments.

Authoring guidance

Start with the narrowest permissions possible. A reviewer usually needs repository reads, not writes or bash.
Keep AGENT.md focused on stable behavior. Put task-specific procedures in skills so the model loads them only when relevant.
Declare every external command in both requirements.bin and the bash allowlist when the agent must execute it.
Test from a directory first with zot inspect and zot run.
Test denied behavior as well as allowed behavior. Ask the agent to attempt an out-of-scope read, write, and command.
Pack the same source twice and compare behavior and output digests before distributing it.
Inspect and verify the final archive, not only its source directory.
Do not claim that network, environment, signatures, extensions, installation, or registry delivery are secured or supported by the local runtime.

Current scope

Implemented now:

local directories and .zot archives
local-only resolution for single-part names
local-first owner/repository and owner/repository/path resolution to any public GitHub repository
temporary execution of agent directories from public GitHub repositories
zot pack, zot inspect, zot verify, and zot run
canonical tar creation with zstd compression
archive digest reporting and safe extraction limits
AGENT.md append or replacement behavior
bundled skills
runtime, OS, and binary requirements
model context and reasoning requirements
filesystem scope enforcement
bash none, ask, and allowlist modes
digest-scoped consent receipts
private agent data directories
agent-scoped sessions

Not implemented yet:

zot install, zot agents, zot use, zot update, or zot publish
installed-name, arbitrary URL, OCI, configurable registry, or zot.sh index resolution
network allowlist enforcement
environment-variable filtering
safely confined bundled executable extensions
automatic artifact theme loading
agent dependency composition
per-agent Telegram, bot, or RPC selection
--no-sandbox, --trust-bash, or registry trust override flags

Treat the unsupported fields and commands in the broader zotfile proposal as forward-looking design, not as current runtime behavior.

source

1# zotfile agents
2
3A zotfile packages an agent's behavior into one portable `.zot` file. It can contain the agent's instructions, reusable skills, static assets, and metadata describing the runtime, model, operating-system, binary, and tool permissions it needs.
4
5The current implementation supports creating, inspecting, verifying, and running local directories and `.zot` archives. It can also run an agent repository or directory directly from any public GitHub repository without keeping a clone. zot has no built-in owner or official collection. Indexed registry distribution, installation, signatures, bundled executable extensions, network permissions, and environment permissions are not implemented yet.
6
7## Quick start
8
9Create a directory for the agent:
10
11```text
12reviewer/
13├── manifest.json
14├── AGENT.md
15└── skills/
16    └── code-review/SKILL.md
17```
18
19Add a minimal `manifest.json`:
20
21```json
22{
23  "zotfile": 1,
24  "name": "code-reviewer",
25  "version": "0.1.0",
26  "description": "Reviews a repository and reports actionable findings.",
27  "runtime": {
28    "min_zot": "0.2.76"
29  },
30  "model": {
31    "requires": ["tools"],
32    "min_context": 64000
33  },
34  "permissions": {
35    "fs": {
36      "read": ["${workspace}"],
37      "write": []
38    },
39    "bash": {
40      "mode": "none"
41    }
42  },
43  "requirements": {
44    "os": ["darwin", "linux", "windows"]
45  },
46  "entry": {
47    "greeting": "What should I review?",
48    "default_prompt": null
49  }
50}
51```
52
53Add the agent's standing instructions to `AGENT.md`:
54
55```markdown
56# Code reviewer
57
58Review the current repository without modifying it.
59
60Prioritize correctness, security, regressions, and missing tests. Report findings
61in severity order with file and line references. Do not praise the code or
62summarize files that have no findings.
63```
64
65Test the directory directly during development:
66
67```bash
68zot inspect ./reviewer
69zot run ./reviewer
70zot run ./reviewer "Review the authentication package"
71```
72
73Package it when ready:
74
75```bash
76zot pack ./reviewer
77# wrote code-reviewer.zot
78# digest sha256:...
79
80zot verify ./code-reviewer.zot
81zot run ./code-reviewer.zot
82```
83
84The first run displays the declared permissions and asks for consent before the agent starts.
85
86## Directory layout
87
88```text
89my-agent/
90├── manifest.json          # identity, requirements, and permissions, required
91├── AGENT.md               # persona and standing instructions
92├── skills/                # SKILL.md directories, optional
93│   └── my-skill/SKILL.md
94├── assets/                # static files, optional
95└── README.md              # human-facing documentation, optional
96```
97
98Only `manifest.json`. Other ordinary files and directories are included in the archive, but the runtime gives no special behavior to `assets/` or `README.md` yet. Files in `assets/` are not automatically added to the model context or granted filesystem access.
99
100Do not include executable extensions in `extensions/`. The local runtime rejects any bundled extension directory containing `extension.json` because extension subprocesses cannot yet be confined to the manifest permissions. This is a deliberate fail-closed restriction.
101
102Symlinks are not supported and cause `zot pack` to fail.
103
104## `AGENT.md`
105
106`AGENT.md` is optional. When present, it is appended to zot's system prompt by default. It should describe the agent's role, workflow, constraints, output format, and when to load bundled skills. When absent, the system prompt is left unchanged.
107
108Keep capability and security declarations out of this file. Permissions come only from `manifest.json` and are enforced independently of agent-authored prose.
109
110Set `replace_system_prompt` to `true` in the manifest to use `AGENT.md` as the replacement system prompt:
111
112```json
113{
114  "replace_system_prompt": true
115}
116```
117
118Replacement is intended for fully specialized agents. The default layering behavior is usually preferable because it retains zot's normal identity and tool-use guidance. With replacement enabled, an empty `AGENT.md` intentionally omits the built-in identity while retaining generated context.
119
120## Bundled skills
121
122Place skills under `skills/<name>/SKILL.md` using the normal zot skill format:
123
124```text
125skills/
126└── investigate-failure/
127    └── SKILL.md
128```
129
130```markdown
131---
132name: investigate-failure
133description: Diagnose a failing command from its output and relevant source files.
134---
135
136# Investigate failure
137
1381. Reproduce the failure if permissions allow it.
1392. Identify the first meaningful error rather than downstream noise.
1403. Read the implementation and nearest tests.
1414. Propose the smallest correction and validation plan.
142```
143
144Bundled skills are added to normal skill discovery while the zotfile is running. The model sees their name and description in the skill manifest and can load the full body through the `skill` tool. See [skills.md](skills.md) for the complete skill format.
145
146## Manifest reference
147
148The current manifest shape is:
149
150```json
151{
152  "zotfile": 1,
153  "name": "code-reviewer",
154  "version": "0.1.0",
155  "description": "Reviews a repository and reports actionable findings.",
156  "license": "MIT",
157  "runtime": {
158    "min_zot": "0.2.76"
159  },
160  "model": {
161    "requires": ["tools", "reasoning"],
162    "min_context": 64000,
163    "preferred": ["claude-opus-4-5", "gpt-5.5"],
164    "min_tier": ""
165  },
166  "permissions": {
167    "fs": {
168      "read": ["${workspace}", "${agent_data}"],
169      "write": ["${agent_data}"]
170    },
171    "bash": {
172      "mode": "allowlist",
173      "allow": ["git", "go"]
174    },
175    "net": {
176      "allow": []
177    },
178    "env": {
179      "read": []
180    }
181  },
182  "requirements": {
183    "bin": ["git", "go"],
184    "os": ["darwin", "linux"]
185  },
186  "entry": {
187    "greeting": "What should I review?",
188    "default_prompt": null
189  },
190  "replace_system_prompt": false
191}
192```
193
194### Top-level fields
195
196| Field | Required | Current behavior |
197|---|---:|---|
198| `zotfile` | yes | Format version. Must be `1`. |
199| `name` | yes | Local agent identity and session/data namespace. See naming rules below. |
200| `version` | no | Displayed in consent and inspection output. Semver is recommended but not currently validated. |
201| `description` | no | Displayed by `zot inspect`. |
202| `license` | no | Package metadata. Not otherwise interpreted by the local runtime. |
203| `runtime.min_zot` | no | Minimum zot version. Older binaries refuse to run the agent. Unversioned development builds cannot satisfy a non-empty minimum. |
204| `model` | no | Model capabilities, minimum context, and preferences. |
205| `permissions` | no | Filesystem and bash permission ceiling. Omitted scopes deny access. |
206| `requirements` | no | Required operating systems and executables. |
207| `entry` | no | Initial presentation and prompt metadata. |
208| `replace_system_prompt` | no | Replaces rather than appends to the normal system prompt when `true`. Defaults to `false`. |
209
210### Name rules
211
212The local runtime currently accepts a flat lowercase name containing only:
213
214- `a` to `z`
215- `0` to `9`
216- `.`, `-`, and `_`
217
218Examples:
219
220```text
221code-reviewer
222repo_onboarder
223acme.audit
224```
225
226Names containing `/`, uppercase letters, spaces, or path-like values such as `..` are rejected. Registry-style `namespace/agent` names are reserved for the future distribution runtime and are not accepted locally yet.
227
228### Model requirements
229
230`model.requires` currently recognizes:
231
232| Capability | Behavior |
233|---|---|
234| `tools` | Accepted. Every model in the current catalog is treated as tool-capable. |
235| `reasoning` | Requires a catalog model marked as supporting reasoning. |
236| `vision` | Recognized, but currently fails closed because vision support is not represented in the model catalog. No model qualifies. |
237
238Unknown capabilities are rejected. Use `tools`, not the draft spelling `tool_calling`.
239
240`model.min_context` requires a catalog model with at least that many context tokens.
241
242`model.preferred` is an ordered list of model IDs. It is a preference, not a provider requirement. Selection proceeds as follows:
243
2441. An explicitly selected model, if present, must satisfy the requirements.
2452. The user's configured default is retained if it satisfies them.
2463. The first compatible model in `preferred` is selected.
2474. Otherwise, zot chooses a compatible active catalog model.
2485. If none qualifies, execution stops with an error.
249
250`model.min_tier` is present in the format shape but is not supported by the local runtime. Leave it empty or omit it. A non-empty value is rejected.
251
252### Runtime and host requirements
253
254Use `requirements.os` to restrict the agent to Go operating-system names:
255
256```json
257{
258  "requirements": {
259    "os": ["darwin", "linux", "windows"]
260  }
261}
262```
263
264Use `requirements.bin` for commands that must already be on `PATH`:
265
266```json
267{
268  "requirements": {
269    "bin": ["git", "python3"]
270  }
271}
272```
273
274Zot checks these requirements before requesting consent. These zotfiles do not have install or postinstall hooks, so authors must document how users can obtain missing programs.
275
276### Entry fields
277
278`entry.pre` is auto-submitted once when `zot run` starts, before the initial prompt is applied. Values that begin with `!` use the interactive shell-escape path (or `BashTool` in `--print` / `--stream` / `--json`). Shell output streams live to the TUI (interactive) or stderr (non-interactive). Other values are sent as a normal user turn; in `--stream` mode that turn's assistant text also streams to stdout. After `pre` finishes, zot reloads extensions and rediscovers skills so anything installed by that command is available for the following turn.
279
280```json
281{
282  "entry": {
283    "pre": "!npx -y skills add tt-a1i/archify --skill archify --agent cursor --copy --yes"
284  }
285}
286```
287
288`entry.default_prompt` is used when the user runs the agent without a prompt. It may be a string or `null`:
289
290```json
291{
292  "entry": {
293    "default_prompt": "Review the current repository"
294  }
295}
296```
297
298In interactive mode, when no prompt is given on the command line, `default_prompt` pre-fills the editor after `pre` finishes (or immediately when `pre` is empty). It is not auto-submitted. A prompt supplied on the command line takes precedence the same way.
299
300In `--print` / `--stream` / `--json` modes, `pre` still runs first: shell escapes abort the run on failure; plain-text `pre` is sent as a first agent turn before the main prompt (`default_prompt` or the CLI prompt).
301
302`entry.greeting` is part of the manifest, but the current local runtime does not render it yet.
303
304## Filesystem permissions
305
306Filesystem permissions are enforced by the built-in `read`, `write`, and `edit` tools. Empty or omitted scopes deny the operation.
307
308Two variables are available in filesystem scopes:
309
310| Variable | Resolves to |
311|---|---|
312| `${workspace}` | The current working directory used for the run. |
313| `${agent_data}` | The agent's persistent private data directory under `$ZOT_HOME/agents/<name>/data/`. |
314
315Relative paths are resolved beneath `${workspace}`. For example:
316
317```json
318{
319  "permissions": {
320    "fs": {
321      "read": ["src", "go.mod"],
322      "write": ["${agent_data}"]
323    }
324  }
325}
326```
327
328This allows reads under `<workspace>/src` and at `<workspace>/go.mod`, plus writes under the private agent data directory. It does not grant general workspace writes.
329
330Common permission profiles follow.
331
332Read-only repository agent:
333
334```json
335{
336  "permissions": {
337    "fs": {
338      "read": ["${workspace}"],
339      "write": []
340    },
341    "bash": { "mode": "none" }
342  }
343}
344```
345
346Repository editor with private persistent state:
347
348```json
349{
350  "permissions": {
351    "fs": {
352      "read": ["${workspace}", "${agent_data}"],
353      "write": ["${workspace}", "${agent_data}"]
354    },
355    "bash": { "mode": "none" }
356  }
357}
358```
359
360Data-only agent:
361
362```json
363{
364  "permissions": {
365    "fs": {
366      "read": ["${agent_data}"],
367      "write": ["${agent_data}"]
368    },
369    "bash": { "mode": "none" }
370  }
371}
372```
373
374Filesystem checks canonicalize existing paths, or the nearest existing parent for a new path, so symlinks cannot be used to escape a declared scope.
375
376If the working directory changes during a run, `${workspace}` permissions are expanded again for the new directory.
377
378## Bash permissions
379
380`permissions.bash.mode` supports three values.
381
382### `none`
383
384```json
385{
386  "bash": { "mode": "none" }
387}
388```
389
390All bash tool calls are denied. This is also the default when `mode` is omitted.
391
392### `ask`
393
394```json
395{
396  "bash": { "mode": "ask" }
397}
398```
399
400Bash is available after the user approves the agent at launch. Consent for an agent using `ask` is requested on every launch and is not cached.
401
402In the current local runtime, `ask` is launch-time capability consent. It does not open a separate confirmation dialog for every individual bash invocation.
403
404The optional Windows `powershell` tool also requires `bash.mode: "ask"` for packaged agents. It is denied in all other permission modes and while jailed. Bash allowlists cannot safely validate PowerShell syntax. Enable the tool via the Windows-only **PowerShell tool** toggle in `/settings` or select it explicitly with `--tools read,write,edit,glob,powershell`; see [PowerShell tool](../README.md#powershell-tool-windows).
405
406### `allowlist`
407
408```json
409{
410  "bash": {
411    "mode": "allowlist",
412    "allow": ["git", "go", "grep"]
413  }
414}
415```
416
417Every command in a shell expression must be listed. Zot checks commands separated by `;`, `&&`, `||`, and pipes. Paths are reduced to their base command name, so `/usr/bin/git` is checked as `git`. A leading environment assignment is skipped when identifying the command.
418
419The allowlist rejects shell substitution, newlines, redirection, backticks, `$()`, `<`, and `>`. This intentionally supports simple command pipelines rather than arbitrary shell programs.
420
421An `allowlist` mode with an empty `allow` array is invalid.
422
423Only declare commands the agent genuinely needs. `requirements.bin` checks that a command exists, while `permissions.bash.allow` controls whether the agent may invoke it. Agents that execute a required command normally need it in both places.
424
425## Network and environment permissions
426
427The manifest shape includes `permissions.net.allow` and `permissions.env.read`, but the local runtime does not enforce them yet. To avoid presenting unenforced declarations as security controls, it rejects manifests containing any network host or environment variable.
428
429Use empty arrays or omit these sections:
430
431```json
432{
433  "permissions": {
434    "net": { "allow": [] },
435    "env": { "read": [] }
436  }
437}
438```
439
440Bundled executable extensions are rejected for the same reason. Network filtering, scrubbed extension environments, and extension confinement belong to a later runtime milestone.
441
442## Consent and persistent data
443
444Before the first run, zot prints the agent identity and expanded filesystem and bash permissions, then asks:
445
446```text
447Agent code-reviewer@0.1.0 wants to run.
448
449  fs read: /path/to/workspace
450  fs write: none
451  bash: none
452
453Allow? [y/N]
454```
455
456For modes other than `bash: ask`, approval is cached for the exact artifact digest under:
457
458```text
459$ZOT_HOME/agents/<name>/consents/<digest>.json
460```
461
462Any change to the packaged artifact produces a different digest and requires consent again. `bash: ask` always requires fresh launch consent.
463
464Non-interactive runs refuse to bypass consent by default. For controlled automation, pass `-y` / `--yes` or set:
465
466```bash
467zot run -y ./code-reviewer.zot --print "Review this repository"
468```
469
470```bash
471ZOT_AGENT_CONSENT=1 zot run ./code-reviewer.zot --print "Review this repository"
472```
473
474This environment variable skips the consent prompt. Only use it after independently inspecting and trusting the exact artifact being run.
475
476Persistent agent data lives under:
477
478```text
479$ZOT_HOME/agents/<name>/data/
480```
481
482The directory is created for every run, but the agent can access it only when `${agent_data}` appears in the relevant filesystem permission scope.
483
484## Agent-scoped sessions
485
486Sessions created by a zotfile are isolated from ordinary zot sessions and from other agents:
487
488```text
489$ZOT_HOME/sessions/agents/<name>/
490```
491
492Normal session flags still apply, including `--continue`, `--resume`, `--session`, and `--no-session`, but their default storage root is scoped to the active agent. The interactive `/sessions` picker and `/session` import, fork, and tree commands use that same agent-scoped root.
493
494## Commands
495
496### `zot run`
497
498Run an unpackaged directory during development:
499
500```bash
501zot run ./my-agent
502zot run -y ./my-agent
503zot run ./my-agent "Do the task"
504zot run ./my-agent -y --print "Do the task"
505```
506
507Run a packed artifact:
508
509```bash
510zot run ./my-agent.zot
511zot run ./my-agent.zot --print "Do the task"
512zot run ./my-agent.zot --stream "Do the task"
513zot run ./my-agent.zot --json "Do the task"
514```
515
516Arguments after the reference use zot's normal CLI parser, so model, provider, reasoning, cwd, session, tool, and output-mode flags remain available. The manifest still imposes its permission and compatibility ceiling. In interactive mode, enabling **show loaded resources at startup** in `/settings` displays the manifest name in an `[Agent]` section above the transcript.
517
518Local filesystem directories, local archive paths, GitHub shorthand, and public GitHub agent-directory URLs are accepted. A single-part name resolves only to a matching local directory or `.zot` archive:
519
520```bash
521zot run my-agent
522```
523
524A two-part name is treated as a GitHub `owner/repository` reference. Resolution remains local-first, then uses the repository root:
525
526```bash
527zot run frkr/zot-archify
528```
529
530Use three or more parts to select an agent directory from any public GitHub collection:
531
532```bash
533zot run acme/agents/reviewer
534zot run acme/clankers/reviewers/go
535```
536
537There is no built-in owner, official collection, collection configuration, or allowlist. Every shorthand part must use lowercase letters, digits, dots, hyphens, or underscores. Backslashes are also accepted as separators on Windows. Prefix a reference with `./`, use an absolute path, retain the `.zot` suffix, or provide a complete GitHub URL to force local or explicit remote resolution.
538
539For GitHub, zot downloads the repository archive into a temporary directory, selects the requested agent subdirectory, validates it, runs it, and removes the downloaded files when the command exits:
540
541```bash
542zot run https://github.com/acme/agents/reviewer
543```
544
545A standard GitHub tree URL is also supported:
546
547```bash
548zot run https://github.com/acme/agents/tree/main/reviewer
549```
550
551Short forms read the repository's default branch through GitHub's `HEAD` archive. A tree URL uses the branch or tag in the URL. Private repositories and branch or tag names containing `/` are not currently supported. The downloaded source is temporary, but normal agent data, consent receipts, and session transcripts remain under `$ZOT_HOME`.
552
553`owner/repository` and `owner/repository/path` are direct GitHub mappings, not an indexed or signed registry. Installed names, arbitrary URLs, and OCI references are not resolved yet.
554
555### `zot pack`
556
557```bash
558zot pack [directory] [output]
559```
560
561Examples:
562
563```bash
564zot pack
565zot pack ./my-agent
566zot pack ./my-agent ./dist/reviewer.zot
567```
568
569The default directory is the current directory. When output is omitted, zot uses `<manifest-name>.zot`. If the output has no `.zot` suffix, zot adds one.
570
571Packing validates the manifest and directory, then creates a zstd-compressed tar archive. Entries are sorted and tar metadata is normalized with fixed timestamps and numeric ownership. The output archive itself is excluded when it is located inside the source directory. Zot prints the SHA-256 digest of the resulting compressed artifact.
572
573### `zot inspect`
574
575```bash
576zot inspect ./my-agent
577zot inspect ./my-agent.zot
578zot inspect https://github.com/acme/agents/code-reviewer
579```
580
581Inspection validates and prints the agent's name, version, description, digest, declared permissions, and complete file list. It does not execute the agent.
582
583For a directory, the digest is computed from its canonical uncompressed tar representation. For a `.zot` archive, the digest is computed from the archive bytes. Therefore a source-directory digest and its packed-archive digest are not expected to match.
584
585### `zot verify`
586
587```bash
588zot verify ./my-agent.zot
589```
590
591The local `verify` command validates that the archive can be safely loaded, validates the manifest and required layout, and prints its SHA-256 digest.
592
593Despite the command name, it does **not** verify a publisher signature in the current runtime. Signature and namespace verification are planned for registry distribution.
594
595## Packaging and extraction limits
596
597The archive loader applies these limits:
598
599| Limit | Value |
600|---|---:|
601| Compressed `.zot` file | 100 MiB |
602| One extracted entry | 64 MiB |
603| Total extracted content | 256 MiB |
604
605Archive extraction rejects absolute paths and parent traversal. Unsupported tar entry types are ignored. New packages use zstd; gzip-compressed tar archives are accepted only as a compatibility fallback for earlier experiments.
606
607## Authoring guidance
608
6091. Start with the narrowest permissions possible. A reviewer usually needs repository reads, not writes or bash.
6102. Keep `AGENT.md` focused on stable behavior. Put task-specific procedures in skills so the model loads them only when relevant.
6113. Declare every external command in both `requirements.bin` and the bash allowlist when the agent must execute it.
6124. Test from a directory first with `zot inspect` and `zot run`.
6135. Test denied behavior as well as allowed behavior. Ask the agent to attempt an out-of-scope read, write, and command.
6146. Pack the same source twice and compare behavior and output digests before distributing it.
6157. Inspect and verify the final archive, not only its source directory.
6168. Do not claim that network, environment, signatures, extensions, installation, or registry delivery are secured or supported by the local runtime.
617
618## Current scope
619
620Implemented now:
621
622- local directories and `.zot` archives
623- local-only resolution for single-part names
624- local-first `owner/repository` and `owner/repository/path` resolution to any public GitHub repository
625- temporary execution of agent directories from public GitHub repositories
626- `zot pack`, `zot inspect`, `zot verify`, and `zot run`
627- canonical tar creation with zstd compression
628- archive digest reporting and safe extraction limits
629- `AGENT.md` append or replacement behavior
630- bundled skills
631- runtime, OS, and binary requirements
632- model context and reasoning requirements
633- filesystem scope enforcement
634- bash `none`, `ask`, and `allowlist` modes
635- digest-scoped consent receipts
636- private agent data directories
637- agent-scoped sessions
638
639Not implemented yet:
640
641- `zot install`, `zot agents`, `zot use`, `zot update`, or `zot publish`
642- installed-name, arbitrary URL, OCI, configurable registry, or zot.sh index resolution
643- network allowlist enforcement
644- environment-variable filtering
645- safely confined bundled executable extensions
646- automatic artifact theme loading
647- agent dependency composition
648- per-agent Telegram, bot, or RPC selection
649- `--no-sandbox`, `--trust-bash`, or registry trust override flags
650
651Treat the unsupported fields and commands in the broader zotfile proposal as forward-looking design, not as current runtime behavior.

zotfiles.md RawHistoryFolder