configure

Name: configure
Author: aviflombaum

38stars2forks

How to install configure?

aviflombaum/claude-code-in-avinyc/configure

$npx -y skills add aviflombaum/claude-code-in-avinyc --skill configure

Installs into the current project.

Summary

Configure or reconfigure qmd collections for this project. Triggers on "configure qmd", "set up qmd", "reconfigure qmd", "qmd setup".

Files · 1

View on GitHub

SKILL.md

1	# QMD Configure
2
3	Interactive interview to set up or reconfigure qmd collections for a project. Idempotent: works for both first-time setup and reconfiguration.
4
5	Run this flow step by step. Do NOT skip steps or assume answers.
6
7	## Step 1: Check qmd is installed
8
9	```bash
10	command -v qmd
11	```
12
13	If missing, tell the user:
14
15	> qmd is not installed. Install it with: `npm install -g @tobilu/qmd`
16
17	Then STOP.
18
19	## Step 2: Check qmd MCP server
20
21	The search skill uses qmd's MCP tools for best performance (models stay warm between queries). Check if the MCP server is configured:
22
23	```bash
24	cat .mcp.json 2>/dev/null \| grep -q '"qmd"' && echo "found in project" \|\| echo "not in project"
25	```
26
27	If not configured, tell the user:
28
29	> qmd MCP server is not configured. Add it for the best search experience.
30	>
31	> Option 1 — CLI (recommended):
32	> ```bash
33	> claude mcp add qmd -- qmd mcp
34	> ```
35	>
36	> Option 2 — Manual: Add to `.mcp.json` in the project root:
37	> ```json
38	> {
39	> "mcpServers": {
40	> "qmd": {
41	> "command": "qmd",
42	> "args": ["mcp"]
43	> }
44	> }
45	> }
46	> ```
47	>
48	> The search skill will fall back to CLI if MCP isn't available, but MCP is significantly faster for repeated queries.
49
50	Continue with setup regardless — MCP is recommended but not required.
51
52	## Step 3: Derive project name
53
54	Get the git repo folder name and normalize it:
55
56	```bash
57	basename "$(git rev-parse --show-toplevel)" \| tr '[:upper:]' '[:lower:]' \| sed 's/[^a-z0-9]/_/g; s/__*/_/g; s/^_//; s/_$//'
58	```
59
60	Show the derived name to the user via `AskUserQuestion`. Let them confirm or type a custom name. Once confirmed, this is the canonical project name used for all collection prefixes.
61
62	## Step 4: Scan for indexable directories
63
64	Find directories containing markdown files:
65
66	```bash
67	find . -maxdepth 2 -type f -name "*.md" -exec dirname {} \; \| sort -u
68	```
69
70	From the output, select only directories useful for a searchable document collection (docs/, plans/, tasks/, etc.). Ignore tmp/, vendor/, node_modules/, and other project cruft.
71
72	## Step 5: Ask which directories to index
73
74	Use `AskUserQuestion` with `multiSelect: true`. List indexable directories as options. Explain that each directory becomes a qmd collection for fast semantic search.
75
76	If no candidate directories were found, ask the user to type custom directory paths.
77
78	## Step 6: Get collection details
79
80	For each selected directory, use `AskUserQuestion` to:
81	- Ask for a short description of what the directory contains (e.g., "Project architecture and feature docs")
82	- Confirm the file pattern (default: `*/.md`, offer alternatives like `*/.{md,txt}`)
83
84	## Step 7: Add collections
85
86	For each selected directory, derive the collection name: `{project}_{dirname}` where dirname has `/` and `.` replaced with `_` and leading dots stripped. Example: `.cursor/rules` → `cursor_rules`, so collection is `myproject_cursor_rules`.
87
88	Get the absolute path to the directory:
89
90	```bash
91	echo "$(git rev-parse --show-toplevel)/<dirname>"
92	```
93
94	Then add the collection:
95
96	```bash
97	qmd collection add "<absolute_path>" --name "<collection_name>" --mask "<pattern>"
98	qmd context add "qmd://<collection_name>/" "<description>"
99	```
100
101	If `qmd collection add` fails (collection already exists), ask the user whether to overwrite. If yes:
102
103	```bash
104	qmd collection remove "<collection_name>"
105	qmd collection add "<absolute_path>" --name "<collection_name>" --mask "<pattern>"
106	qmd context add "qmd://<collection_name>/" "<description>"
107	```
108
109	## Step 8: Write project config
110
111	Use the Write tool to create `.claude/qmd.json`:
112
113	```json
114	{
115	"project": "<project_name>",
116	"collections": {
117	"<project>_<dirname>": {
118	"path": "<relative-dir-path>",
119	"pattern": "*/.md",
120	"description": "<user-provided description>"
121	}
122	},
123	"gitHook": false
124	}
125	```
126
127	## Step 9: Generate embeddings
128
129	```bash
130	qmd update && qmd embed
131	```
132
133	Tell the user it's generating embeddings. This may take a moment for large collections.
134
135	## Step 10: Ask about git post-commit hook
136
137	Use `AskUserQuestion`: "Install a git post-commit hook? When enabled, committing changes to .md files will automatically re-index in the background so search results stay fresh."
138
139	Options: "Yes, install hook" / "No, skip"
140
141	If yes, install the hook by appending to `.git/hooks/post-commit` (create the file if needed, ensure it's executable):
142
143	```bash
144	# qmd-auto-index:<project_name>
145	# Auto-update qmd index when markdown files change
146	export PATH="$HOME/.bun/bin:$HOME/.local/bin:$PATH"
147	if command -v qmd &>/dev/null; then
148	if git diff-tree --no-commit-id --name-only -r HEAD \| grep -q '\.md$'; then
149	(qmd update && qmd embed) &>/dev/null &
150	fi
151	fi
152	```
153
154	Check for the marker comment `# qmd-auto-index:<project_name>` first to avoid duplicates. Update `.claude/

People who liked this also viewed

6 picks

Type

hyperframes-cliUse the HyperFrames CLI development loop: init, add, catalog, capture, lint, check, snapshot, compare, grade-compare, preview, play, present, beats, keyframes, single or batch render, publish, cloud, SkillsJul 2026262k37k
request-refactor-planSkills for Real Engineers. Straight from my .agents directory.SkillsJul 2026165k181k
analyze-projectRigor Analyze / Rigor Audit read-only skill for deep learning research repositories. Use when the user wants to read and understand a repository, inspect model structure and training or inference entrSkillsMay 2026161k507
explore-codeRigor Improve implementation leaf skill for auditable candidate implementation in deep learning research repositories. Use when the researcher explicitly authorizes exploratory work on an isolated braSkillsMay 2026161k507
ubiquitous-languageSkills for Real Engineers. Straight from my .agents directory.SkillsJul 2026160k181k
scaffold-exercisesSkills for Real Engineers. Straight from my .agents directory.SkillsJul 2026149k181k

configure

38stars2forks

View on GitHub

How to install configure?

aviflombaum/claude-code-in-avinyc/configure

$npx -y skills add aviflombaum/claude-code-in-avinyc --skill configure

Installs into the current project.

Summary

Configure or reconfigure qmd collections for this project. Triggers on "configure qmd", "set up qmd", "reconfigure qmd", "qmd setup".

Files · 1

View on GitHub

SKILL.md

1	# QMD Configure
2
3	Interactive interview to set up or reconfigure qmd collections for a project. Idempotent: works for both first-time setup and reconfiguration.
4
5	Run this flow step by step. Do NOT skip steps or assume answers.
6
7	## Step 1: Check qmd is installed
8
9	```bash
10	command -v qmd
11	```
12
13	If missing, tell the user:
14
15	> qmd is not installed. Install it with: `npm install -g @tobilu/qmd`
16
17	Then STOP.
18
19	## Step 2: Check qmd MCP server
20
21	The search skill uses qmd's MCP tools for best performance (models stay warm between queries). Check if the MCP server is configured:
22
23	```bash
24	cat .mcp.json 2>/dev/null \| grep -q '"qmd"' && echo "found in project" \|\| echo "not in project"
25	```
26
27	If not configured, tell the user:
28
29	> qmd MCP server is not configured. Add it for the best search experience.
30	>
31	> Option 1 — CLI (recommended):
32	> ```bash
33	> claude mcp add qmd -- qmd mcp
34	> ```
35	>
36	> Option 2 — Manual: Add to `.mcp.json` in the project root:
37	> ```json
38	> {
39	> "mcpServers": {
40	> "qmd": {
41	> "command": "qmd",
42	> "args": ["mcp"]
43	> }
44	> }
45	> }
46	> ```
47	>
48	> The search skill will fall back to CLI if MCP isn't available, but MCP is significantly faster for repeated queries.
49
50	Continue with setup regardless — MCP is recommended but not required.
51
52	## Step 3: Derive project name
53
54	Get the git repo folder name and normalize it:
55
56	```bash
57	basename "$(git rev-parse --show-toplevel)" \| tr '[:upper:]' '[:lower:]' \| sed 's/[^a-z0-9]/_/g; s/__*/_/g; s/^_//; s/_$//'
58	```
59
60	Show the derived name to the user via `AskUserQuestion`. Let them confirm or type a custom name. Once confirmed, this is the canonical project name used for all collection prefixes.
61
62	## Step 4: Scan for indexable directories
63
64	Find directories containing markdown files:
65
66	```bash
67	find . -maxdepth 2 -type f -name "*.md" -exec dirname {} \; \| sort -u
68	```
69
70	From the output, select only directories useful for a searchable document collection (docs/, plans/, tasks/, etc.). Ignore tmp/, vendor/, node_modules/, and other project cruft.
71
72	## Step 5: Ask which directories to index
73
74	Use `AskUserQuestion` with `multiSelect: true`. List indexable directories as options. Explain that each directory becomes a qmd collection for fast semantic search.
75
76	If no candidate directories were found, ask the user to type custom directory paths.
77
78	## Step 6: Get collection details
79
80	For each selected directory, use `AskUserQuestion` to:
81	- Ask for a short description of what the directory contains (e.g., "Project architecture and feature docs")
82	- Confirm the file pattern (default: `*/.md`, offer alternatives like `*/.{md,txt}`)
83
84	## Step 7: Add collections
85
86	For each selected directory, derive the collection name: `{project}_{dirname}` where dirname has `/` and `.` replaced with `_` and leading dots stripped. Example: `.cursor/rules` → `cursor_rules`, so collection is `myproject_cursor_rules`.
87
88	Get the absolute path to the directory:
89
90	```bash
91	echo "$(git rev-parse --show-toplevel)/<dirname>"
92	```
93
94	Then add the collection:
95
96	```bash
97	qmd collection add "<absolute_path>" --name "<collection_name>" --mask "<pattern>"
98	qmd context add "qmd://<collection_name>/" "<description>"
99	```
100
101	If `qmd collection add` fails (collection already exists), ask the user whether to overwrite. If yes:
102
103	```bash
104	qmd collection remove "<collection_name>"
105	qmd collection add "<absolute_path>" --name "<collection_name>" --mask "<pattern>"
106	qmd context add "qmd://<collection_name>/" "<description>"
107	```
108
109	## Step 8: Write project config
110
111	Use the Write tool to create `.claude/qmd.json`:
112
113	```json
114	{
115	"project": "<project_name>",
116	"collections": {
117	"<project>_<dirname>": {
118	"path": "<relative-dir-path>",
119	"pattern": "*/.md",
120	"description": "<user-provided description>"
121	}
122	},
123	"gitHook": false
124	}
125	```
126
127	## Step 9: Generate embeddings
128
129	```bash
130	qmd update && qmd embed
131	```
132
133	Tell the user it's generating embeddings. This may take a moment for large collections.
134
135	## Step 10: Ask about git post-commit hook
136
137	Use `AskUserQuestion`: "Install a git post-commit hook? When enabled, committing changes to .md files will automatically re-index in the background so search results stay fresh."
138
139	Options: "Yes, install hook" / "No, skip"
140
141	If yes, install the hook by appending to `.git/hooks/post-commit` (create the file if needed, ensure it's executable):
142
143	```bash
144	# qmd-auto-index:<project_name>
145	# Auto-update qmd index when markdown files change
146	export PATH="$HOME/.bun/bin:$HOME/.local/bin:$PATH"
147	if command -v qmd &>/dev/null; then
148	if git diff-tree --no-commit-id --name-only -r HEAD \| grep -q '\.md$'; then
149	(qmd update && qmd embed) &>/dev/null &
150	fi
151	fi
152	```
153
154	Check for the marker comment `# qmd-auto-index:<project_name>` first to avoid duplicates. Update `.claude/

People who liked this also viewed

6 picks

Type