agentsmesh 0.8.0 → 0.10.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/CHANGELOG.md CHANGED
@@ -1,5 +1,71 @@
1
1
  # Changelog
2
2
 
3
+ ## 0.10.0
4
+
5
+ ### Minor Changes
6
+
7
+ - c4fb261: Add `kilo-code` as a new built-in target. Kilo Code is a multi-surface AI coding platform (VS Code extension, JetBrains plugin, CLI, cloud agent) and a fork of Roo Code (which is a fork of Cline).
8
+
9
+ Generation always uses Kilo's new layout: `AGENTS.md` (root), `.kilo/rules/`, `.kilo/commands/`, `.kilo/agents/` (first-class subagents), `.kilo/skills/`, `.kilo/mcp.json`, and `.kilocodeignore`. Import covers BOTH the new layout and Kilo's legacy layout (`.kilocode/`, `.kilocodemodes`) so existing kilo / Roo-era projects round-trip cleanly.
10
+
11
+ Capabilities (project + global):
12
+ - `rules`, `additionalRules`, `commands`, `agents`, `skills`, `mcp`, `ignore`: native
13
+ - `hooks`: none — Kilo Code has no user-facing lifecycle hook system; canonical hooks emit a lint warning.
14
+ - `permissions`: none — Kilo permissions live in `kilo.jsonc`, which agentsmesh does not generate in v1; canonical permissions emit a lint warning.
15
+
16
+ Global mode generates under `~/.kilo/` (`AGENTS.md`, `rules/`, `commands/`, `agents/`, `skills/`, `mcp.json`) plus `~/.kilocodeignore`, and mirrors skills into `~/.agents/skills/` for cross-tool compatibility (suppressed when `codex-cli` is also active).
17
+
18
+ Use `agentsmesh import --from kilo-code` to migrate existing Kilo projects (new or legacy layout) into canonical `.agentsmesh/`, then `agentsmesh generate --targets kilo-code` to project them back as the documented new layout.
19
+
20
+ - 5d6cfbb: Sequential `agentsmesh import --from <target>` runs now merge MCP servers by name into `.agentsmesh/mcp.json` instead of replacing the whole file. Existing canonical entries are preserved and the imported set wins on name collision, so a `claude-code` import followed by a `cursor` import keeps both targets' servers in canonical state.
21
+
22
+ Affects every importer that writes `mcp.json`: `claude-code` (`.claude/settings.json` + `.mcp.json` + `~/.claude/.mcp.json`), `codex-cli` (`config.toml`), `continue`, `cursor`, and any descriptor-driven importer using `mode: 'mcpJson'`. The previous behavior — last import overwrites the file and silently drops earlier servers — is gone.
23
+
24
+ Also fixed: a build-time regression where `writeMcpWithMerge` was referenced by five importers without the backing module being shipped, breaking `tsc --noEmit` for consumers building from source.
25
+
26
+ ## 0.9.0
27
+
28
+ ### Minor Changes
29
+
30
+ - b3f702d: Adds three new `agentsmesh lint` diagnostics, a recommended `.gitignore` policy for materialized packs, and a one-step `agentsmesh target-scaffold` workflow.
31
+
32
+ Added:
33
+ - New lint diagnostics, all emitted as warnings (do not change `lint`'s exit code):
34
+ - `silent-drop-guard` flags canonical content a target would otherwise drop without trace based on its capability map.
35
+ - `hook-script-references` warns when a hook command references a script path for any target whose generator does not copy hook scripts into its output. **All built-in targets except Copilot fall under this rule today.** Existing hook configs that reference local script paths (e.g. `./scripts/pre-commit.sh`) will surface a new warning per matching target on the first lint after upgrade. The script must already exist relative to the hook execution directory or the generated config will fail at runtime — the diagnostic just makes the gap visible.
36
+ - `rule-scope-inversion` catches manual-activation rules whose scope contradicts the target's projection rules.
37
+ All three are wired into `runLint` for every target via descriptor capabilities; no existing rule has been removed and no diagnostic is upgraded to error severity.
38
+ - `agentsmesh init` now writes `.agentsmesh/packs/` into `.gitignore` alongside `agentsmesh.local.yaml`, `.agentsmeshcache`, and `.agentsmesh/.lock.tmp`, and skips entries already covered by a broader pre-existing pattern (e.g. an existing `.agentsmesh/` line covers `.agentsmesh/packs/`). Materialized packs are treated like `node_modules` — `installs.yaml` is committed and `agentsmesh install --sync` reproduces the tree deterministically post-clone.
39
+ - `agentsmesh target-scaffold` post-steps collapse the previous three-edit sequence into one `pnpm catalog:generate` invocation, backed by a new auto-discovered builtin-target catalog (`scripts/generate-target-catalog.ts` + `pnpm catalog:verify` drift guard in CI).
40
+
41
+ Changed:
42
+ - The `agentsmesh.json` and `pack.json` schemas now list `targets` enums alphabetically. Schema consumers that pin order will see a one-time diff; values are unchanged.
43
+ - README documents the commit-vs-gitignore convention for generated tool folders and clarifies native Windows support (no WSL).
44
+
45
+ Internal:
46
+ - Per-target importers (`antigravity`, `claude-code`, `continue`, `copilot`, `cursor`, `gemini-cli`, `junie`, `kiro`, `roo-code`) migrated to the descriptor-driven import runner with mapper functions extracted into `import-mappers.ts` to keep `index.ts` ↔ `importer.ts` cycles out of the TDZ.
47
+ - New shared link-format registry (`src/core/reference/link-format-registry.ts`) consolidates per-target link rendering rules.
48
+ - `writeFileAtomic` now refuses to follow a pre-existing symlink at the target path (closes a TOCTOU window where a swapped symlink could redirect writes outside the project tree).
49
+ - `agentsmesh plugin add` now warns that plugins load as trusted Node.js modules with full process privileges.
50
+
51
+ ### Patch Changes
52
+
53
+ - 08ef1b0: Security hardening and correctness fixes across install, generate, reference rewriting, plugin loading, and caching subsystems.
54
+
55
+ Fixed:
56
+ - `agentsmesh install --path` now rejects paths that traverse outside the install source root, closing a directory-escape vulnerability where `--path ../../outside` could read files outside the fetched source.
57
+ - Pack names are validated as single directory segments before materialization — values containing path separators (e.g. `../../escape`) are rejected, preventing writes outside `.agentsmesh/packs/`.
58
+ - `writeFileAtomic` now checks the `.tmp` staging path for symlinks before writing, closing a TOCTOU window where a symlink at `${path}.tmp` could redirect writes to an attacker-controlled location.
59
+ - `agentsmesh generate --targets` now validates filter values against configured targets and errors on unknown names. Previously a misspelled target silently produced zero outputs and `--check` reported success, allowing CI to pass while checking nothing.
60
+ - Project-scope skill mirrors now receive source-map entries for reference rewriting. Previously only global-scope mirrors were mapped, leaving Markdown links inside project-mirrored skill files unrewritten.
61
+ - Plugin targets declaring `sharedArtifacts` are now recognized during global reference rewriting. Previously only builtin target descriptors were consulted, so plugin-owned shared paths could be rebased through the wrong artifact map.
62
+ - `runDescriptorImport` is now exported from `agentsmesh/targets` as documented in the plugin guide.
63
+ - Importer fallback sources are now tried when configured primary files are absent on disk, not only when the primary source list is empty.
64
+ - `flatFile` and `mcpJson` import modes now honor `canonicalDir` when `canonicalFilename` is a bare filename, matching the documented directory-plus-filename contract for plugin descriptor authors.
65
+ - Rule path mapping uses POSIX `basename` with backslash normalization instead of host `node:path`, preventing broken slugs when Windows-shaped canonical paths appear on a POSIX host.
66
+ - Relative `file:` plugin sources now resolve against `projectRoot` instead of the filesystem root.
67
+ - Remote cache keys now preserve dots and use double-separator delimiters so distinct refs like `v1.0.0` and `v1_0_0` no longer collide. Existing cached entries will be re-fetched once after upgrade.
68
+
3
69
  ## 0.8.0
4
70
 
5
71
  ### Minor Changes
package/README.md CHANGED
@@ -1,6 +1,8 @@
1
1
  <div align="center">
2
2
 
3
- # AgentsMesh — AI Coding Config Sync for Every Tool
3
+ # AgentsMesh — One `.agentsmesh/` Directory for Every AI Coding Tool
4
+
5
+ <img src="https://raw.githubusercontent.com/sampleXbro/agentsmesh/master/assets/agentsmesh-banner.jpeg" alt="AgentsMesh — One source. Every AI coding tool. Always in sync." width="100%" />
4
6
 
5
7
  [![CI](https://github.com/sampleXbro/agentsmesh/actions/workflows/ci.yml/badge.svg)](https://github.com/sampleXbro/agentsmesh/actions/workflows/ci.yml)
6
8
  [![npm version](https://img.shields.io/npm/v/agentsmesh.svg)](https://www.npmjs.com/package/agentsmesh)
@@ -12,99 +14,189 @@
12
14
  [![Docs](https://img.shields.io/badge/docs-website-brightgreen.svg)](https://samplexbro.github.io/agentsmesh)
13
15
  [![PRs Welcome](https://img.shields.io/badge/PRs-welcome-brightgreen.svg)](https://github.com/sampleXbro/agentsmesh/pulls)
14
16
 
15
- **AgentsMesh is an open-source CLI and TypeScript library for AI coding configuration sync. One canonical `.agentsmesh/` directory manages rules, prompts, commands, agents, skills, MCP servers, hooks, ignore files, and permissions across every major AI coding tool.**
17
+ </div>
16
18
 
17
- Edit once and generate `CLAUDE.md`, `AGENTS.md`, `.cursor/rules`, `.github/copilot-instructions.md`, `.gemini/settings.json`, `.windsurf/rules`, `.codex/config.toml`, `.kiro/steering`, and more from the same source. Import existing Claude Code, Cursor, Copilot, Gemini CLI, Windsurf, Codex CLI, and other configs back into canonical form without losing round-trip metadata.
19
+ AI coding assistants now ship with their own configuration formats — `CLAUDE.md`, `AGENTS.md`, `.cursor/rules/*.mdc`, `.github/copilot-instructions.md`, `.gemini/settings.json`, `.windsurf/rules/*.md`, `.codex/config.toml`, `.kiro/steering/*.md`, and more. Maintaining the same rules, prompts, MCP servers, hooks, and permissions across all of them by hand causes config drift fast.
18
20
 
19
- **Works with** Claude Code · Cursor · GitHub Copilot · Gemini CLI · Windsurf · Continue · Cline · Kiro · Codex CLI · Junie · Roo Code · Antigravity plus plugin targets. See the [full feature matrix](https://samplexbro.github.io/agentsmesh/reference/supported-tools/).
20
-
21
- </div>
21
+ **AgentsMesh** is an open-source CLI and TypeScript library that fixes this. You write canonical rules, commands, agents, skills, MCP, hooks, ignore files, and permissions once in `.agentsmesh/`, then `agentsmesh generate` projects them out as native config for every supported assistant. `agentsmesh import` brings existing tool configs back into canonical form, and `agentsmesh check` catches drift in CI.
22
22
 
23
23
  > **Full documentation: [samplexbro.github.io/agentsmesh](https://samplexbro.github.io/agentsmesh)**
24
24
 
25
25
  ---
26
26
 
27
- ## Why developers use AgentsMesh
27
+ ## Before / After
28
+
29
+ **Before — fragmented assistant-native config in one repo:**
30
+
31
+ ```text
32
+ CLAUDE.md
33
+ AGENTS.md
34
+ .cursor/rules/*.mdc
35
+ .github/copilot-instructions.md
36
+ .gemini/settings.json
37
+ .windsurf/rules/*.md
38
+ .codex/config.toml
39
+ .kiro/steering/*.md
40
+ ```
41
+
42
+ **After — one canonical source, generated everywhere:**
43
+
44
+ ```text
45
+ .agentsmesh/
46
+ rules/
47
+ _root.md
48
+ commands/
49
+ agents/
50
+ skills/
51
+ mcp.json
52
+ hooks.yaml
53
+ permissions.yaml
54
+ ignore
55
+ ```
28
56
 
29
- - **Unify AI coding rules** across Claude Code, Cursor, Copilot, Gemini CLI, Windsurf, Codex CLI, and mixed-tool teams.
30
- - **Adopt existing projects safely** with bidirectional `import` and `generate` instead of rewriting every tool config by hand.
31
- - **Sync personal global config** from `~/.agentsmesh/` to user-level assistant folders such as `~/.claude/`, `~/.cursor/`, and `~/.codex/`.
32
- - **Standardize MCP, hooks, permissions, skills, and agents** where tools support them natively, with metadata-backed projections where they do not.
33
- - **Catch config drift in CI** with lock-file checks, diffs, linting, and merge recovery built for team workflows.
34
- - **Share and extend configuration** with community packs, remote `extends`, runtime plugins, and a typed programmatic API.
57
+ ```bash
58
+ npx agentsmesh generate
59
+ ```
60
+
61
+ The native files above are still emitted AgentsMesh writes them for you from `.agentsmesh/`. Edit canonical sources, regenerate, and every tool stays in sync.
35
62
 
36
63
  ---
37
64
 
38
- ## Install
65
+ ## 60-second quickstart
39
66
 
40
- Requires **Node.js 20+**. Supported platforms: **Linux**, **macOS**, and **Windows** (native, not WSL).
67
+ Requires Node.js 20+. Works on Linux, macOS, and Windows (native, not WSL).
41
68
 
42
69
  ```bash
43
- npm install -D agentsmesh # or: pnpm add -D / yarn add -D
44
- npx agentsmesh --help # run without installing
70
+ npx agentsmesh init # scaffold .agentsmesh/ + agentsmesh.yaml
71
+ npx agentsmesh generate # produce native configs for every enabled tool
72
+ npx agentsmesh check # CI-friendly drift gate against .agentsmesh/.lock
45
73
  ```
46
74
 
47
- CLI aliases: `agentsmesh` and `amsh`.
75
+ - **`init`** — creates `agentsmesh.yaml`, `agentsmesh.local.yaml`, and the canonical `.agentsmesh/` directory.
76
+ - **`generate`** — writes `CLAUDE.md`, `AGENTS.md`, `.cursor/`, `.github/copilot-instructions.md`, etc. from canonical sources.
77
+ - **`check`** — exits non-zero if generated files have drifted from `.agentsmesh/.lock`. Drop into CI.
48
78
 
49
- > **Windows notes:** All paths are normalized internally so generated configs and `installs.yaml` are portable across platforms. Watch mode uses polling on Windows because `ReadDirectoryChangesW` can miss just-created files in tmpdirs. CI runs the full test suite on Linux, macOS, and Windows.
79
+ Prefer a local install? `npm install -D agentsmesh` (also `pnpm add -D` / `yarn add -D`). The CLI ships as `agentsmesh` and the shorter alias `amsh`.
50
80
 
51
81
  ---
52
82
 
53
- ## Quick start
83
+ ## Safe adoption in an existing repository
54
84
 
55
- ### New project
85
+ If your repo already has `.cursor/`, `.claude/`, `.github/copilot-instructions.md`, or other native files, you don't have to delete them. The recommended flow imports them into `.agentsmesh/` first, lets you preview the projection, and only then trusts `generate`.
56
86
 
57
87
  ```bash
58
- agentsmesh init # scaffold .agentsmesh/
59
- # edit .agentsmesh/rules/_root.md
60
- agentsmesh generate # produce configs for every enabled tool
88
+ npx agentsmesh import --from cursor # or claude-code, copilot, codex-cli, gemini-cli, windsurf, ...
89
+ npx agentsmesh diff # patch-style preview of what generate would change
90
+ npx agentsmesh generate # write native configs (back) from canonical
91
+ npx agentsmesh check # add to CI to detect drift
61
92
  ```
62
93
 
63
- ### Existing project — adopt with one import
94
+ What this gets you:
64
95
 
65
- ```bash
66
- agentsmesh import --from cursor # or claude-code, copilot, codex-cli, ...
67
- agentsmesh generate
68
- ```
96
+ - `import` reads existing tool configs and writes equivalent canonical files into `.agentsmesh/` — round-trip metadata is preserved so re-import doesn't lose information.
97
+ - `diff` shows the unified patch every output file would receive, so you can review before any write.
98
+ - `check` reads `.agentsmesh/.lock` and fails the build if the canonical sources and the generated files disagree.
99
+
100
+ `import --from` accepts the built-in target IDs: `claude-code`, `cursor`, `copilot`, `codex-cli`, `gemini-cli`, `windsurf`, `continue`, `cline`, `kiro`, `kilo-code`, `junie`, `roo-code`, `antigravity`. Plugin targets are valid too.
101
+
102
+ ---
103
+
104
+ ## Demo
105
+
106
+ <!-- TODO: Add terminal demo GIF showing init → generate → diff → check. -->
69
107
 
70
- ### Personal global config
108
+ A quick sample of the canonical → native projection:
71
109
 
72
110
  ```bash
73
- agentsmesh init --global
74
- agentsmesh import --global --from claude-code
75
- agentsmesh generate --global # writes to ~/.claude/, ~/.cursor/, etc.
111
+ npx agentsmesh init
112
+ find .agentsmesh -maxdepth 2 -type f # see the canonical scaffold
113
+ npx agentsmesh generate
114
+ npx agentsmesh diff # preview future changes
115
+ npx agentsmesh check # CI-style drift gate
76
116
  ```
77
117
 
118
+ On macOS/Linux you can also run `tree .agentsmesh` if you have `tree` installed.
119
+
120
+ ---
121
+
122
+ ## Supported AI coding tools
123
+
124
+ AgentsMesh currently generates native config for every major AI coding assistant — Claude Code, Cursor, GitHub Copilot, Gemini CLI, Windsurf, Continue, Cline, Kiro, Codex CLI, Junie, Roo Code, Antigravity — plus plugin targets you can ship as standalone npm packages. Each tool's native vs. embedded support per feature is tracked in the [supported tools matrix](https://samplexbro.github.io/agentsmesh/reference/supported-tools/). The full matrix table is also embedded [further down this README](#supported-tools--feature-matrix).
125
+
126
+ ---
127
+
128
+ ## Why developers use AgentsMesh
129
+
130
+ - **Bidirectional sync** — `import` reads existing tool configs into `.agentsmesh/`; `generate` projects them back out. Round-trips are loss-free, so adopting AgentsMesh in an existing repo never throws away data.
131
+ - **Automatic link rebasing** — references like `.agentsmesh/skills/foo/SKILL.md` are rewritten to target-relative paths in every generated artifact, so cross-file links stay valid from `.claude/`, `.cursor/`, `.github/`, `.codex/`, and the rest.
132
+ - **Managed embedding with round-trip metadata** — when a target has no native slot for a feature (e.g. commands in Codex CLI, agents in Cline), AgentsMesh embeds it with frontmatter that survives the next `import`. No silent data loss; the full feature-by-feature breakdown lives in the [supported tools matrix](https://samplexbro.github.io/agentsmesh/reference/supported-tools/).
133
+ - **Team-safe collaboration** — `agentsmesh check` is a CI drift gate against `.agentsmesh/.lock`, `agentsmesh diff` previews changes, `agentsmesh merge` rebuilds the lock after three-way Git conflicts, and `lock_features` + per-feature `strategy` prevent accidental overrides.
134
+ - **Global mode** — `~/.agentsmesh/` syncs personal AI config to `~/.claude/`, `~/.cursor/`, `~/.codex/`, `~/.windsurf/`, and other user-level folders. Every CLI command accepts `--global`.
135
+ - **Extensible** — community packs (`agentsmesh install ...`), remote `extends`, runtime plugins (`agentsmesh plugin add`), schema-validated config files, and a typed programmatic API for scripts, IDE extensions, and CI.
136
+
137
+ ---
138
+
139
+ ## Why not just `AGENTS.md`?
140
+
141
+ [`AGENTS.md`](https://agents.md) is great as a shared, human-readable instruction file. AgentsMesh uses `AGENTS.md` natively where the target supports it (Codex CLI, Cursor, Copilot, Junie, Windsurf, …) and treats it as a first-class output, not a competitor.
142
+
143
+ The reason `AGENTS.md` alone is not enough: most AI coding assistants expose configuration surfaces beyond a single instruction markdown file, and those surfaces don't all overlap.
144
+
145
+ - **Cursor** has `.cursor/rules/*.mdc` (with frontmatter scopes), `.cursorignore`, MCP config, and hooks.
146
+ - **Claude Code** has `CLAUDE.md`, `.claude/agents/`, `.claude/skills/`, `.claude/commands/`, `.claude/settings.json`, hooks, and permissions.
147
+ - **GitHub Copilot** has `.github/copilot-instructions.md`, `.github/instructions/*.instructions.md`, agents, prompts, and (partial) hooks.
148
+ - **Gemini CLI** has `GEMINI.md`, `.gemini/settings.json` (MCP + hooks), `.gemini/commands/*.toml`, and agents.
149
+ - **Codex CLI** has `AGENTS.md` plus `.codex/config.toml`, `.codex/agents/*.toml`, and `.codex/rules/`.
150
+ - **Windsurf**, **Continue**, **Cline**, **Kiro**, **Junie**, **Roo Code**, **Antigravity** each have their own native rules, workflows, MCP servers, skills, and ignore files.
151
+
152
+ AgentsMesh canonicalizes all of these — rules, commands, agents, skills, MCP servers, hooks, ignore patterns, permissions — so you don't pick one tool's surface as the lowest common denominator. When a tool has no native slot for a feature, AgentsMesh embeds it with round-trip metadata instead of dropping it.
153
+
78
154
  ---
79
155
 
80
- ## High-demand features
156
+ ## Core concepts
157
+
158
+ `.agentsmesh/` is the canonical source of truth. Generated tool files are artifacts. The directory contains:
81
159
 
82
- ### AI coding config sync for every tool
160
+ - `rules/_root.md` the root rule every target projects (typically becomes `CLAUDE.md`, `AGENTS.md`, `.cursor/rules/_root.mdc`, etc.).
161
+ - `rules/*.md` — additional scoped rules.
162
+ - `commands/*.md` — reusable slash-style prompts/commands.
163
+ - `agents/*.md` — agent definitions (where the target supports them).
164
+ - `skills/<name>/SKILL.md` (+ supporting files) — composable skills.
165
+ - `mcp.json` — MCP server definitions.
166
+ - `hooks.yaml` — pre/post tool hooks.
167
+ - `permissions.yaml` — allow/deny rules where the target supports them.
168
+ - `ignore` — paths the assistant should not read or modify.
83
169
 
84
- AgentsMesh generates native configuration for every major AI coding assistant. Each tool's files are produced from a single `.agentsmesh/` directory with support for **rules, commands, agents, skills, MCP servers, hooks, ignore patterns, and permissions**:
170
+ Configuration:
85
171
 
86
- | Tool | Main files generated |
87
- |------|---------------------|
88
- | **Claude Code** | `CLAUDE.md`, `.claude/agents/`, `.claude/skills/`, `.claude/commands/`, `.claude/settings.json`, `.claude/hooks.json`, MCP via `.claude.json` |
89
- | **Cursor** | `.cursor/rules/*.mdc`, `AGENTS.md`, `.cursor/mcp.json`, `.cursor/hooks.json`, `.cursorignore` |
90
- | **GitHub Copilot** | `.github/copilot-instructions.md`, `.github/instructions/*.instructions.md`, `.github/agents/`, `.github/prompts/` |
91
- | **Gemini CLI** | `GEMINI.md`, `.gemini/settings.json` (MCP + hooks), `.gemini/commands/*.toml`, `.gemini/agents/` |
92
- | **Windsurf** | `.windsurf/rules/*.md`, `.windsurf/workflows/`, `.windsurf/mcp_config.json`, `.windsurf/hooks.json` |
93
- | **Continue** | `.continue/rules/`, `.continue/prompts/`, `.continue/mcpServers/`, `.continue/config.yaml` |
94
- | **Cline** | `.clinerules/`, `.cline/skills/`, `.cline/cline_mcp_settings.json`, hooks |
95
- | **Kiro** | `.kiro/steering/`, `.kiro/skills/`, `.kiro/hooks/*.kiro.hook`, `.kiro/settings/mcp.json` |
96
- | **Codex CLI** | `AGENTS.md`, `.codex/config.toml`, `.codex/agents/*.toml`, `.codex/rules/` |
97
- | **Junie** | `AGENTS.md`, `.junie/agents/`, `.junie/commands/`, `.junie/skills/`, `.junie/mcp/mcp.json` |
98
- | **Roo Code** | `.roo/rules/`, `.roomodes` (agents → custom modes), `.roo/commands/`, `.roo/skills/` |
99
- | **Antigravity** | `.agents/rules/general.md`, `.agents/skills/`, `.agents/workflows/`, `.agents/mcp_config.json` |
172
+ - `agentsmesh.yaml` selects which targets and features are enabled.
173
+ - `agentsmesh.local.yaml` — per-developer overrides (gitignored by default).
174
+ - `.agentsmesh/.lock` drift-detection lock file consumed by `agentsmesh check`.
100
175
 
101
- When a tool lacks native support for a feature, AgentsMesh embeds it with round-trip metadata — no data loss on re-import. See the [supported tools matrix](https://samplexbro.github.io/agentsmesh/reference/supported-tools/) for per-tool native vs. embedded breakdown.
176
+ Detailed contracts: [Canonical Config](https://samplexbro.github.io/agentsmesh/canonical-config/) · [Generation pipeline](https://samplexbro.github.io/agentsmesh/reference/generation-pipeline/).
102
177
 
103
- ### Bidirectional import and lossless generate
178
+ ---
179
+
180
+ ## CLI usage
181
+
182
+ ```bash
183
+ agentsmesh init [--global] [--yes]
184
+ agentsmesh generate [--global] [--targets <csv>] [--check] [--dry-run] [--force] [--refresh-cache]
185
+ agentsmesh import --from <target> [--global]
186
+ agentsmesh diff [--global] [--targets <csv>]
187
+ agentsmesh lint [--global] [--targets <csv>]
188
+ agentsmesh watch [--global] [--targets <csv>]
189
+ agentsmesh check [--global]
190
+ agentsmesh merge [--global]
191
+ agentsmesh matrix [--global] [--targets <csv>] [--verbose]
192
+ agentsmesh install <source> [--sync] [--path <dir>] [--target <id>] [--as <kind>] [--name <id>] [--extends] [--dry-run] [--global] [--force]
193
+ agentsmesh plugin add|list|remove|info [--version <v>] [--id <id>]
194
+ agentsmesh target scaffold <id> [--name <displayName>] [--force]
195
+ ```
104
196
 
105
- Use `agentsmesh import --from <tool>` to migrate existing AI tool configs into canonical `.agentsmesh/` files, then `agentsmesh generate` to project them back out. Managed embedding, reference rewriting, and lock metadata preserve projected features so commands, agents, and skills can round-trip even when a target stores them differently.
197
+ `agentsmesh --help` prints the same surface; `agentsmesh <cmd> --help` is also supported.
106
198
 
107
- ### Global mode for personal AI assistant config
199
+ ### Global mode (personal AI assistant config)
108
200
 
109
201
  `.agentsmesh/` at the project level is for teams. `~/.agentsmesh/` at the home level is for personal setup across every repo you touch:
110
202
 
@@ -114,7 +206,7 @@ agentsmesh import --global --from claude-code
114
206
  agentsmesh generate --global # writes ~/.claude/CLAUDE.md, ~/.cursor/, ~/.codex/, ~/.windsurf/, etc.
115
207
  ```
116
208
 
117
- Every built-in target and every plugin supports global mode. Every CLI command (`diff`, `lint`, `watch`, `check`, `merge`, `matrix`) accepts `--global`. [Global mode paths per tool →](https://samplexbro.github.io/agentsmesh/reference/supported-tools/#global-mode)
209
+ Every built-in target with a global layout supports global mode. Every CLI command (`diff`, `lint`, `watch`, `check`, `merge`, `matrix`) accepts `--global`. [Global mode paths per tool →](https://samplexbro.github.io/agentsmesh/reference/supported-tools/#global-mode)
118
210
 
119
211
  ### Plugins for new AI coding tools
120
212
 
@@ -130,12 +222,12 @@ Plugins have full parity with built-in targets: project + global layouts, featur
130
222
 
131
223
  ### Team-safe collaboration & CI drift detection
132
224
 
133
- - **`agentsmesh check`** — CI gate that exits 1 if generated files drifted from the lock
134
- - **`agentsmesh diff`** — preview what the next `generate` would change
135
- - **`agentsmesh lint`** — validate canonical config against target-specific constraints
136
- - **`agentsmesh watch`** — regenerate target files on save during local editing
137
- - **`agentsmesh merge`** — recover from three-way `.lock` conflicts after `git merge`
138
- - **Collaboration config** — `lock_features` and `strategy` prevent accidental overrides
225
+ - **`agentsmesh check`** — CI gate that exits 1 if generated files drifted from the lock.
226
+ - **`agentsmesh diff`** — preview what the next `generate` would change.
227
+ - **`agentsmesh lint`** — validate canonical config against target-specific constraints; also surfaces cross-target warnings (`silent-drop-guard`, `hook-script-references`, `rule-scope-inversion`) for content a target would silently drop or mishandle. [Lint reference →](https://samplexbro.github.io/agentsmesh/cli/lint/)
228
+ - **`agentsmesh watch`** — regenerate target files on save during local editing.
229
+ - **`agentsmesh merge`** — recover from three-way `.lock` conflicts after `git merge`.
230
+ - **Collaboration config** — `lock_features` and `strategy` prevent accidental overrides.
139
231
 
140
232
  ### Community packs and shared config
141
233
 
@@ -167,13 +259,6 @@ Why generated configs stay committed: the same reason `package-lock.json` does.
167
259
 
168
260
  If your team has a strong reason to gitignore generated configs (e.g., monorepo size concerns, regenerate-on-checkout hooks), add the target-specific entries manually — but expect to wire `agentsmesh generate` into your post-checkout flow.
169
261
 
170
- ### Extending AgentsMesh
171
-
172
- - **`agentsmesh target scaffold foo-ide`** — generate a built-in target skeleton (10 files: descriptor, generators, importer, linter, tests, fixtures) with global mode, conversion support, and lint hooks pre-wired. The catalog is auto-discovered at build time (`pnpm catalog:generate`) — no manual edits to `target-ids.ts`, `builtin-targets.ts`, or the import-map barrel.
173
- - **`agentsmesh plugin add <pkg>`** — load third-party npm packages as runtime targets with full built-in parity. Supports `agentsmesh plugin list`, `info`, and `remove`.
174
-
175
- [Extending guide →](https://samplexbro.github.io/agentsmesh/guides/extending/) · [Building plugins →](https://samplexbro.github.io/agentsmesh/guides/building-plugins/)
176
-
177
262
  ### Schema-validated configs (IDE autocomplete)
178
263
 
179
264
  Every config file ships with a generated JSON Schema, so VS Code, JetBrains, and other editors give you autocomplete and validation out of the box:
@@ -190,45 +275,7 @@ Every config file ships with a generated JSON Schema, so VS Code, JetBrains, and
190
275
 
191
276
  ---
192
277
 
193
- ## Supported tools feature matrix
194
-
195
- ### Project scope (`agentsmesh generate`)
196
-
197
- <!-- agentsmesh:support-matrix:project -->
198
- | Feature | Antigravity | Claude Code | Cline | Codex CLI | Continue | Copilot | Cursor | Gemini CLI | Junie | Kiro | Roo Code | Windsurf |
199
- |---|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|
200
- | Rules | Native | Native | Native | Native | Native | Native | Native | Native | Native | Native | Native | Native |
201
- | Additional Rules | Native | Native | Native | Native | Native | Native | Embedded | Embedded | Native | Native | Native | Native |
202
- | Commands | Partial (workflows) | Native | Native (workflows) | Embedded | Embedded | Native | Native | Native | Native | — | Native | Native (workflows) |
203
- | Agents | — | Native | Embedded | Native | — | Native | Native | Native | Native | Native | Partial | Embedded |
204
- | Skills | Native | Native | Native | Native | Embedded | Native | Native | Native | Native | Native | Native | Native |
205
- | MCP Servers | — | Native | Native | Native | Native | — | Native | Native | Native | Native | Native | Partial |
206
- | Hooks | — | Native | Native | — | — | Partial | Native | Partial | — | Native | — | Native |
207
- | Ignore | — | Native | Native | — | — | — | Native | Native (settings-embedded) | Native | Native | Native | Native |
208
- | Permissions | — | Native | — | — | — | — | Partial | Partial | — | — | — | — |
209
- <!-- /agentsmesh:support-matrix:project -->
210
-
211
- ### Global scope (`agentsmesh generate --global`)
212
-
213
- <!-- agentsmesh:support-matrix:global -->
214
- | Feature | Antigravity | Claude Code | Cline | Codex CLI | Continue | Copilot | Cursor | Gemini CLI | Junie | Kiro | Roo Code | Windsurf |
215
- |---|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|
216
- | Rules | Native | Native | Native | Native | Native | Native | Native | Native | Native | Native | Native | Native |
217
- | Additional Rules | Embedded | Native | Native | Embedded | Native | Native | Embedded | Embedded | Embedded | Native | Native | Partial |
218
- | Commands | Partial (workflows) | Native | Native (workflows) | Embedded | Native | Native | Native | Native | Native | — | Native | Native (workflows) |
219
- | Agents | — | Native | Embedded | Native | — | Native | Native | Native | Native | Native | Partial | Embedded |
220
- | Skills | Native | Native | Native | Native | Native | Native | Native | Native | Native | Native | Native | Native |
221
- | MCP Servers | Native | Native | Native | Native | Native | — | Native | Native | Native | Native | Native | Partial |
222
- | Hooks | — | Native | Native | — | — | — | Native | Partial | — | — | — | Native |
223
- | Ignore | — | Native | Native | — | — | — | Native | — | — | Native | Native | Native |
224
- | Permissions | — | Native | — | — | — | — | — | — | — | — | — | — |
225
- <!-- /agentsmesh:support-matrix:global -->
226
-
227
- See the [full feature matrix docs](https://samplexbro.github.io/agentsmesh/reference/supported-tools/) for native vs. embedded support details and per-tool global paths.
228
-
229
- ---
230
-
231
- ## Programmatic API
278
+ ## TypeScript / Programmatic API
232
279
 
233
280
  AgentsMesh is also importable as a typed ESM library, so you can drive every CLI capability — `generate`, `import`, `lint`, `diff`, `check` — from scripts, IDE extensions, MCP servers, or CI without spawning the CLI. Public entrypoints: `agentsmesh` (full surface), `agentsmesh/engine`, `agentsmesh/canonical`, `agentsmesh/targets`.
234
281
 
@@ -286,6 +333,44 @@ Every public symbol resolves to a real `.d.ts` under strict TypeScript. Full ref
286
333
 
287
334
  ---
288
335
 
336
+ ## Supported tools — feature matrix
337
+
338
+ ### Project scope (`agentsmesh generate`)
339
+
340
+ <!-- agentsmesh:support-matrix:project -->
341
+ | Feature | Antigravity | Claude Code | Cline | Codex CLI | Continue | Copilot | Cursor | Gemini CLI | Junie | Kilo Code | Kiro | Roo Code | Windsurf |
342
+ |---|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|
343
+ | Rules | Native | Native | Native | Native | Native | Native | Native | Native | Native | Native | Native | Native | Native |
344
+ | Additional Rules | Native | Native | Native | Native | Native | Native | Embedded | Embedded | Native | Native | Native | Native | Native |
345
+ | Commands | Partial (workflows) | Native | Native (workflows) | Embedded | Embedded | Native | Native | Native | Native | Native | — | Native | Native (workflows) |
346
+ | Agents | — | Native | Embedded | Native | — | Native | Native | Native | Native | Native | Native | Partial | Embedded |
347
+ | Skills | Native | Native | Native | Native | Embedded | Native | Native | Native | Native | Native | Native | Native | Native |
348
+ | MCP Servers | — | Native | Native | Native | Native | — | Native | Native | Native | Native | Native | Native | Partial |
349
+ | Hooks | — | Native | Native | — | — | Partial | Native | Partial | — | — | Native | — | Native |
350
+ | Ignore | — | Native | Native | — | — | — | Native | Native (settings-embedded) | Native | Native | Native | Native | Native |
351
+ | Permissions | — | Native | — | — | — | — | Partial | Partial | — | — | — | — | — |
352
+ <!-- /agentsmesh:support-matrix:project -->
353
+
354
+ ### Global scope (`agentsmesh generate --global`)
355
+
356
+ <!-- agentsmesh:support-matrix:global -->
357
+ | Feature | Antigravity | Claude Code | Cline | Codex CLI | Continue | Copilot | Cursor | Gemini CLI | Junie | Kilo Code | Kiro | Roo Code | Windsurf |
358
+ |---|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|
359
+ | Rules | Native | Native | Native | Native | Native | Native | Native | Native | Native | Native | Native | Native | Native |
360
+ | Additional Rules | Embedded | Native | Native | Embedded | Native | Native | Embedded | Embedded | Embedded | Native | Native | Native | Partial |
361
+ | Commands | Partial (workflows) | Native | Native (workflows) | Embedded | Native | Native | Native | Native | Native | Native | — | Native | Native (workflows) |
362
+ | Agents | — | Native | Embedded | Native | — | Native | Native | Native | Native | Native | Native | Partial | Embedded |
363
+ | Skills | Native | Native | Native | Native | Native | Native | Native | Native | Native | Native | Native | Native | Native |
364
+ | MCP Servers | Native | Native | Native | Native | Native | — | Native | Native | Native | Native | Native | Native | Partial |
365
+ | Hooks | — | Native | Native | — | — | — | Native | Partial | — | — | — | — | Native |
366
+ | Ignore | — | Native | Native | — | — | — | Native | — | — | Native | Native | Native | Native |
367
+ | Permissions | — | Native | — | — | — | — | — | — | — | — | — | — | — |
368
+ <!-- /agentsmesh:support-matrix:global -->
369
+
370
+ See the [full feature matrix docs](https://samplexbro.github.io/agentsmesh/reference/supported-tools/) for native vs. embedded support details and per-tool global paths.
371
+
372
+ ---
373
+
289
374
  ## Documentation
290
375
 
291
376
  - **[Getting Started](https://samplexbro.github.io/agentsmesh/getting-started/installation/)** — installation, quick start
@@ -1,5 +1,5 @@
1
- import { b as CanonicalFiles, V as ValidatedConfig } from './schema-BeGiBqbB.js';
2
- export { C as CanonicalAgent, a as CanonicalCommand, c as CanonicalRule, d as CanonicalSkill, H as HookEntry, e as Hooks, I as IgnorePatterns, M as McpConfig, f as McpServer, P as Permissions, S as SkillSupportingFile, g as StdioMcpServer, U as UrlMcpServer } from './schema-BeGiBqbB.js';
1
+ import { b as CanonicalFiles, V as ValidatedConfig } from './schema-qelg8gw8.js';
2
+ export { C as CanonicalAgent, a as CanonicalCommand, c as CanonicalRule, d as CanonicalSkill, H as HookEntry, e as Hooks, I as IgnorePatterns, M as McpConfig, f as McpServer, P as Permissions, S as SkillSupportingFile, g as StdioMcpServer, U as UrlMcpServer } from './schema-qelg8gw8.js';
3
3
  import 'zod';
4
4
 
5
5
  /**