@phren/cli 0.1.53 → 0.2.1

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/skills/docs.md DELETED
@@ -1,170 +0,0 @@
1
- ---
2
- name: docs
3
- description: Update and verify all phren documentation surfaces after code changes.
4
- dependencies:
5
- - git
6
- ---
7
- # phren-docs - Documentation accuracy check and update
8
-
9
- > Keep all phren documentation surfaces in sync with the code after any change.
10
-
11
- Invoke this after adding tools, changing CLI commands, bumping versions, or editing hooks. It walks every documentation surface, checks what's stale, and updates `docs/documentation.html` (the canonical public-facing reference).
12
-
13
- ## When to run
14
-
15
- - After adding, removing, or renaming MCP tools
16
- - After changing CLI commands or their flags
17
- - After bumping the version number
18
- - After changing hook names, events, or behavior
19
- - Before publishing to npm
20
- - When the user says "update the docs" or "check the docs"
21
-
22
- ## Documentation surfaces
23
-
24
- | File | What it contains | What to check |
25
- |------|-----------------|---------------|
26
- | `README.md` | Project overview, install steps, quick-start | Tool count, version badge, CLI examples |
27
- | `CLAUDE.md` | In-repo instructions for Claude | Tool count (heading says "MCP Tools (N)"), CLI command list, key file list |
28
- | `AGENTS.md` | Agent-facing instructions | Tool count, hook table, CLI commands |
29
- | `CONTRIBUTING.md` | Dev setup, build/test commands | `npm run build`, `npm test`, publish steps |
30
- | `CHANGELOG.md` | Release history | Latest version matches `package.json` version |
31
- | `docs/index.html` | GitHub Pages landing page | Install commands, version references, feature list |
32
- | `docs/documentation.html` | Full reference doc (primary surface) | Tool count, CLI commands, hook table, env vars table |
33
- | `docs/api-reference.md` | API / tool reference | Tool signatures, return shapes |
34
- | `docs/architecture.md` | Data-flow diagrams and architecture | File paths, hook names, flow descriptions |
35
- | `docs/faq.md` | Common questions and answers | References to commands, file paths |
36
- | `docs/llms.txt` | Short LLM-friendly summary | Tool count, version, install command |
37
- | `docs/llms-full.txt` | Full LLM-readable reference | Complete tool list, CLI commands, env vars |
38
- | `mcp/README.md` | MCP server README | Tool count, server entry point |
39
- | `vscode-extension/README.md` | VS Code extension README | Feature list, commands |
40
-
41
- ## Step-by-step
42
-
43
- ### 1. Read the source of truth
44
-
45
- The source of truth for counts and signatures lives in the code, not the docs:
46
-
47
- ```bash
48
- # Count exported MCP tools in index.ts
49
- grep -c 'server\.tool(' mcp/src/index.ts
50
-
51
- # Current version
52
- node -p "require('./package.json').version"
53
-
54
- # CLI commands (scan cli.ts / index.ts for subcommand registrations)
55
- grep 'program\.' mcp/src/cli.ts | head -40
56
- ```
57
-
58
- Record: **tool count**, **version**, and the list of top-level CLI subcommands.
59
-
60
- ### 2. Check each documentation surface
61
-
62
- For each surface in the table above:
63
-
64
- 1. Read the file
65
- 2. Find every place it mentions a tool count (e.g. "65 tools", "MCP Tools (65)")
66
- 3. Find every place it mentions the version number
67
- 4. Find every CLI command listing
68
- 5. Find the hook table (event names, what each hook does)
69
- 6. Note any stale references (old file paths, renamed tools, removed commands)
70
-
71
- ### 3. Spot-check critical numbers
72
-
73
- These numbers appear in multiple files and must all agree:
74
-
75
- | Item | Where to verify |
76
- |------|----------------|
77
- | MCP tool count | `mcp/src/index.ts`: count `server.tool(` calls |
78
- | CLI subcommand count | `mcp/src/cli.ts`: count `program.command(` calls |
79
- | Version | `package.json` → `version` field |
80
- | Hook events | `mcp/src/init.ts`: look for hook registration |
81
-
82
- Run `/parity` if it is installed to automate numeric cross-checking across surfaces.
83
-
84
- ### 4. Update `docs/documentation.html`
85
-
86
- This is the primary public reference. It must reflect the current state of the code.
87
-
88
- Sections to verify and update:
89
-
90
- - **Tool reference table**: every tool name, description, and parameter signature
91
- - **CLI commands block**: every `phren <subcommand>` with flags and description
92
- - **Hooks table**: event name, what it runs, what it does
93
- - **Environment variables table**: variable name, default, description
94
- - **Version number** in the page title, install command snippets, and any badges
95
- - **Tool count** in any summary sentence (e.g. "65 MCP tools")
96
-
97
- When updating the HTML:
98
- - Keep existing structure and CSS classes. Do not restructure the page
99
- - Update text content only; do not rewrite layout
100
- - Preserve the `<details>` blocks for env var categories if they exist
101
-
102
- ### 5. Update `docs/llms.txt` and `docs/llms-full.txt`
103
-
104
- These files are consumed by LLMs directly. Keep them plain text with no HTML.
105
-
106
- `llms.txt`, short summary (under 60 lines):
107
- - Tool count
108
- - Install command
109
- - One-line description of what phren does
110
-
111
- `llms-full.txt`, comprehensive reference:
112
- - All MCP tool signatures with descriptions
113
- - All CLI commands
114
- - All environment variables
115
- - Directory structure
116
-
117
- ### 6. Update `CLAUDE.md` (in-repo)
118
-
119
- The heading `## MCP Tools (N)` must match the actual tool count. Update N if it changed.
120
-
121
- The CLI commands block must list every top-level command. Add or remove lines to match.
122
-
123
- ### 7. Sync version references
124
-
125
- `CHANGELOG.md`: the top entry's version must match `package.json`. If a new version was bumped but the changelog has no entry yet, note it. Do not fabricate one.
126
-
127
- `README.md` and `docs/index.html`: update any hardcoded version strings (badges, `npm install @phren/cli@X.Y.Z`, etc.).
128
-
129
- ### 8. Report
130
-
131
- After reviewing and updating, output:
132
-
133
- ```
134
- phren-docs
135
-
136
- Version: 0.0.8
137
- MCP tools: 65
138
- CLI subcommands: 22
139
-
140
- Surfaces checked: 14
141
-
142
- Changes made:
143
- - docs/documentation.html: updated tool count (64→65), added new env var PHREN_X
144
- - CLAUDE.md: updated MCP Tools heading (64→65)
145
- - docs/llms-full.txt: added PHREN_X to env vars section
146
-
147
- No changes needed:
148
- - README.md: version and tool count already correct
149
- - CHANGELOG.md: top entry matches package.json version
150
- - mcp/README.md: tool count correct
151
-
152
- Warnings:
153
- - docs/faq.md: references `phren init`, command was renamed to `phren link` in v0.0.7
154
- ```
155
-
156
- List every file checked. Be specific about what changed and what line/section.
157
-
158
- ## What not to do
159
-
160
- - Do not restructure or reformat documentation for style. Only fix accuracy
161
- - Do not add new sections or features to the docs without user direction
162
- - Do not fabricate changelog entries for unreleased versions
163
- - Do not silently skip a surface because it looks "probably fine". Check all of them
164
- - Do not guess tool counts. Always derive from `grep` on the source file
165
-
166
- ## Related skills
167
-
168
- - `/parity`: numeric cross-check across all documentation surfaces
169
- - `/consolidate`: surface cross-project patterns before updating architecture docs
170
- - `/discover`: find gaps in project documentation
@@ -1,216 +0,0 @@
1
- ---
2
- name: init
3
- description: Set up a new project in phren with summary, CLAUDE.md, task, and skill templates.
4
- dependencies:
5
- - git
6
- ---
7
- # phren-init - Scaffold a new project
8
-
9
- > Set up a new project in phren with summary, CLAUDE.md, task, and skill templates.
10
-
11
- Tell phren about a new project, or bootstrap phren himself if you're starting fresh.
12
-
13
- ## Usage
14
-
15
- ```
16
- phren-init my-new-project # scaffold a specific project
17
- phren-init # list existing projects, suggest unconfigured ones
18
- ```
19
-
20
- ## First: find or create the phren directory
21
-
22
- ```bash
23
- PHREN_DIR="${PHREN_DIR:-$HOME/.phren}"
24
- ls "$PHREN_DIR" 2>/dev/null
25
- ```
26
-
27
- If the phren directory doesn't exist, offer to create it:
28
- > "No phren repo found. Want me to create one at ~/.phren? This will set up the base directory structure (global/, profiles/, machines.yaml)."
29
-
30
- If the user says yes, create the base structure:
31
- ```bash
32
- mkdir -p "$PHREN_DIR"/{global/skills,profiles}
33
- # Create a starter machines.yaml
34
- # Create a starter profile
35
- ```
36
-
37
- Then continue with the project scaffolding below.
38
-
39
- ## With a project name
40
-
41
- ### 1. Check it doesn't already exist
42
-
43
- ```bash
44
- ls "$PHREN_DIR/<project-name>/" 2>/dev/null
45
- ```
46
-
47
- If it exists, tell the user and ask if they want to reset it or just update specific files.
48
-
49
- ### 2. Create the project directory
50
-
51
- ```bash
52
- mkdir -p "$PHREN_DIR/<project-name>/skills"
53
- ```
54
-
55
- ### 3. Create summary.md
56
-
57
- This is the project's identity card. Five lines, no more.
58
-
59
- ```markdown
60
- # <project-name>
61
-
62
- What: <one sentence about what this project does>
63
- Stack: <languages, frameworks, key deps>
64
- Status: <active / maintenance / new / archived>
65
- Run: <the main command to start it, e.g. "npm run dev">
66
- Watch out: <the one thing that trips people up>
67
- ```
68
-
69
- Ask the user for these if you can't figure them out from the project directory. If the project directory exists on disk (e.g. `~/<project-name>/`), read its package.json, pyproject.toml, README, or similar to pre-fill.
70
-
71
- ### 4. Create CLAUDE.md
72
-
73
- ```markdown
74
- # <project-name>
75
-
76
- <One paragraph: what this project is and what it does.>
77
-
78
- ## Commands
79
-
80
- ```
81
- <dev command>
82
- <build command>
83
- <test command>
84
- ```
85
-
86
- ## Architecture
87
-
88
- <!-- Fill this in as the project grows -->
89
-
90
- ## Conventions
91
-
92
- <!-- Project-specific rules, patterns, naming conventions -->
93
- ```
94
-
95
- Pre-fill from the project's existing README or package.json if available. Leave architecture and conventions as placeholders with the HTML comments.
96
-
97
- ### 5. Create tasks.md
98
-
99
- ```markdown
100
- # <project-name> tasks
101
-
102
- ## Active
103
-
104
- ## Queue
105
-
106
- ## Done
107
- ```
108
-
109
- ### 6. Create skills README
110
-
111
- ```markdown
112
- # <project-name> skills
113
-
114
- Project-specific skills go here. Add `.md` files and phren will resolve them together with global skills, then mirror the effective set into the project's `.claude/skills/` directory on sync.
115
- ```
116
-
117
- Write this to `$PHREN_DIR/<project-name>/skills/README.md`.
118
-
119
- ### 7. Add to profile(s)
120
-
121
- Check if profiles exist:
122
- ```bash
123
- ls "$PHREN_DIR/profiles/"*.yaml 2>/dev/null
124
- ```
125
-
126
- If profiles exist, ask: "Which profile should this project be in?" and show the options.
127
-
128
- If no profiles exist, offer to create one:
129
- > "No profiles set up yet. Want me to create a 'default' profile with this project in it?"
130
-
131
- Then add the project name to the chosen profile's `projects` list.
132
-
133
- ### 8. Commit (if git repo)
134
-
135
- ```bash
136
- cd "$PHREN_DIR"
137
- git add <project-name>/
138
- git add profiles/ # if modified
139
- git commit -m "add <project-name>"
140
- git push # only if remote exists
141
- ```
142
-
143
- If phren isn't a git repo yet, walk them through the full setup:
144
-
145
- ```bash
146
- cd "$PHREN_DIR"
147
- git init
148
- git add -A
149
- git commit -m "initial phren setup"
150
- ```
151
-
152
- Then check if `gh` is available:
153
-
154
- ```bash
155
- which gh && gh auth status
156
- ```
157
-
158
- If yes, offer to create the GitHub repo:
159
- > "Your phren isn't on GitHub yet. Want me to create a private repo and push it? That's what lets you sync across machines."
160
-
161
- If they say yes:
162
- ```bash
163
- gh repo create my-phren --private --source=. --push
164
- ```
165
-
166
- If `gh` isn't available, show the manual steps:
167
- > "Create a private repo at github.com/new, then:
168
- > git remote add origin git@github.com:YOU/my-phren.git
169
- > git push -u origin main
170
- >
171
- > Once it's on GitHub, clone it on any machine and run `phren-sync` to activate."
172
-
173
- ### 9. Report
174
-
175
- ```
176
- phren-init <project-name>
177
-
178
- Created:
179
- $PHREN_DIR/<project-name>/summary.md
180
- $PHREN_DIR/<project-name>/CLAUDE.md
181
- $PHREN_DIR/<project-name>/tasks.md
182
- $PHREN_DIR/<project-name>/skills/README.md
183
-
184
- Added to profile: default
185
-
186
- Run phren-sync to activate on this machine.
187
- ```
188
-
189
- ## Without a project name
190
-
191
- When user just runs `phren-init` with no args:
192
-
193
- 1. If phren directory exists, list all directories that have a `summary.md` (configured projects)
194
- 2. Look for directories in `~/` that look like projects (have package.json, pyproject.toml, Cargo.toml, go.mod, etc.) but don't have a phren entry
195
- 3. Show both lists and suggest adding unconfigured ones
196
-
197
- ```
198
- phren-init: project scan
199
-
200
- Configured in phren:
201
- myapp, api-server
202
-
203
- Found on this machine but not in phren:
204
- ~/side-project (has package.json)
205
- ~/experiments/rust-thing (has Cargo.toml)
206
-
207
- Run phren-init <name> to add one.
208
- ```
209
-
210
- If phren doesn't exist at all, offer to bootstrap it from scratch.
211
-
212
- ## Related skills
213
-
214
- - `phren-sync`: activate the new project on this machine after init
215
- - `phren-discover`: find what to build next in a project
216
- - `/tasks`: manage the project's task queue
@@ -1,121 +0,0 @@
1
- ---
2
- name: profiles
3
- description: Manage machine-to-profile and profile-to-project mappings in phren.
4
- dependencies:
5
- - git
6
- ---
7
- # phren-profiles - Manage your profiles
8
-
9
- > **Multi-machine only.** This skill is for users who sync phren across multiple machines and need to control which projects appear where. If you only use one machine, you can skip this -- `phren-sync` handles everything you need.
10
-
11
- > Add projects to profiles, move projects between profiles, create new profiles, and check what's in each one.
12
-
13
- Manage which projects appear on which machines by editing your profile definitions. Each machine maps to one profile, and each profile is a list of projects.
14
-
15
- ## Prerequisites
16
-
17
- You need a phren repo. The profiles live in `~/.phren/profiles/` and machine mappings live in `~/.phren/machines.yaml`.
18
-
19
- ```
20
- ~/.phren/
21
- profiles/
22
- work.yaml
23
- personal.yaml
24
- machines.yaml
25
- ```
26
-
27
- If you don't have a phren repo yet, start with `phren-init`.
28
-
29
- ## What's a profile?
30
-
31
- A profile is a YAML file listing projects for a machine role. Example:
32
-
33
- ```yaml
34
- name: work
35
- description: Work laptop with company projects
36
- projects:
37
- - global
38
- - my-api
39
- - frontend
40
- ```
41
-
42
- When you sync on a machine mapped to this profile, only those projects appear. Personal projects stay off work machines.
43
-
44
- ## How to use it
45
-
46
- ### "Add a project to a profile"
47
-
48
- User says: "add my-project to my work profile"
49
-
50
- 1. Find the profile file: `~/.phren/profiles/work.yaml`
51
- 2. Read it and find the `projects:` list
52
- 3. Add the project name if it's not already there
53
- 4. Commit the change
54
-
55
- ### "Create a new profile"
56
-
57
- User says: "create a new profile called staging"
58
-
59
- 1. Create a new file: `~/.phren/profiles/staging.yaml`
60
- 2. Fill it with the required structure:
61
-
62
- ```yaml
63
- name: staging
64
- description: Staging servers and experiments
65
- projects:
66
- - global
67
- ```
68
-
69
- 3. Ask which projects should be in this new profile (can add more later)
70
- 4. Commit the file
71
-
72
- ### "Show me what's in a profile"
73
-
74
- User says: "what's in my personal profile?"
75
-
76
- 1. Find `~/.phren/profiles/personal.yaml`
77
- 2. Read it and display the projects list
78
-
79
- ### "Move a project between profiles"
80
-
81
- User says: "move my-project from personal to work"
82
-
83
- 1. Find `~/.phren/profiles/personal.yaml` and remove the project
84
- 2. Find `~/.phren/profiles/work.yaml` and add the project
85
- 3. Commit both changes
86
-
87
- ### "Set this machine's profile"
88
-
89
- User says: "this machine is my work laptop"
90
-
91
- 1. Get the current machine name: `cat ~/.phren/.machine-id` or `hostname`
92
- 2. Ask which profile to use (show available profiles from `~/.phren/profiles/`)
93
- 3. Add/update the line in `~/.phren/machines.yaml`: `work-laptop: work`
94
- 4. Commit the change
95
- 5. Run `phren-sync` to activate
96
-
97
- ### "List my profiles"
98
-
99
- If the user asks what profiles exist:
100
-
101
- 1. List all files in `~/.phren/profiles/`
102
- 2. For each one, read the name and description fields
103
- 3. Show which machine is mapped to each profile (from `machines.yaml`)
104
-
105
- ## After making changes
106
-
107
- Always commit to the phren git repo:
108
-
109
- ```bash
110
- cd ~/.phren
111
- git add profiles/ machines.yaml
112
- git commit -m "update profiles"
113
- git push # only if remote exists
114
- ```
115
-
116
- Then suggest running `phren-sync` to activate changes on this machine.
117
-
118
- ## Related skills
119
-
120
- - `phren-sync`: sync your profiles to this machine and activate them
121
- - `phren-init`: create a new project or bootstrap phren from scratch