dx-roaster

Name: dx-roaster
Author: Varnan-Tech

13installs59forks

How to install dx-roaster?

varnan-tech/opendirectory/dx-roaster

$npx -y skills add Varnan-Tech/opendirectory --skill dx-roaster

Installs into the current project.

Summary

Brutally honest developer-experience audit for a GitHub repo. Scores 10 DX dimensions (time-to-first-success, README clarity, visual proof, install, quick-start, docs, examples, community, trust, marketing), writes a shareable roast in the requested tone (brutal/honest/kind), pro

Files · 1

View on GitHub

SKILL.md

1	# dx-roaster
2
3	Roasts a GitHub repo's developer experience. Produces four files: `roast.md`, `action-plan.md`, `ideal-readme.md`, `score.json`. Built for OSS maintainers and dev-tool founders preparing for Show HN, Product Hunt, or any launch where the first five minutes matter.
4
5	---
6
7	## Critical rules (read before every run)
8
9	1. Read all three reference files BEFORE scoring. `references/scoring-rubric.md`, `references/roast-voice-guide.md`, `references/readme-template.md`. The rubric is non-negotiable — use the exact 10 dimensions.
10
11	2. Score every dimension on the 0–10 scale. Never skip. Missing data = score 0, not "N/A".
12
13	3. Roast must cite specific findings. Quote bad lines verbatim. Reference file paths. Cite line numbers when possible. Never write generic platitudes like "documentation could be better".
14
15	4. Tone matters. Default `honest`. Switch voice per `references/roast-voice-guide.md`. Brutal is funny but never personal. Kind is encouraging but still specific.
16
17	5. Action plan is impact-ranked, not severity-ranked. Use `impact × (1 / effort)`. A 5-minute fix that adds 10 points ranks above a 2-hour fix that adds 12 points.
18
19	6. Never invent URLs. When writing `ideal-readme.md`, use `<placeholder>` syntax for missing links. Roast will tell maintainer to fill in.
20
21	7. Never name maintainers or contributors. Roast the work, not the people.
22
23	8. Output exactly four files in `dx-roast/`. Do not write to other locations. Do not skip files.
24
25	---
26
27	## Step 1: Intake
28
29	Required:
30	- `repo` — local path (default `.`) OR `https://github.com/owner/repo` URL
31
32	Optional:
33
34	\| Parameter \| Default \| Description \|
35	\|---\|---\|---\|
36	\| `tone` \| `honest` \| `brutal` / `honest` / `kind` \|
37	\| `output_dir` \| `dx-roast/` \| Where to write the 4 output files \|
38	\| `compare_to` \| none \| Optional second repo URL for benchmark comparison \|
39
40	If `repo` is a URL: clone to a temp dir using `git clone <url> /tmp/dx-roast-<short-hash>`. Scan from there. Delete temp dir after run.
41
42	If `repo` is `.` or local path: scan in place, do not modify anything.
43
44	---
45
46	## Step 2: Discover
47
48	Read these files (each may be absent — note absence):
49	- `README.md` — full contents, structure (headings), code blocks, image refs, badge URLs
50	- `CONTRIBUTING.md`, `CODE_OF_CONDUCT.md`
51	- `.github/ISSUE_TEMPLATE/`, `.github/PULL_REQUEST_TEMPLATE.md`, `.github/CODEOWNERS`
52	- `package.json` (description, homepage, repository, license, scripts), `pyproject.toml`, `Cargo.toml`, `go.mod`
53	- `LICENSE` (existence + type)
54	- `docs/` folder contents (just file names + sizes)
55	- `examples/` or `samples/` or `cookbook/` folder
56	- `.github/workflows/*.yml` (just names + brief CI presence check)
57
58	If `repo` is a URL, also fetch via `gh`:
59	```bash
60	gh api repos/<owner>/<repo> --jq '{stars: .stargazers_count, forks: .forks_count, watchers: .subscribers_count, pushed_at, description, homepage, topics, default_branch}'
61	gh api repos/<owner>/<repo>/commits --jq '.[0].commit.author.date' \| head -1
62	gh api repos/<owner>/<repo>/contributors --paginate --jq '[.[] \| .login] \| length'
63	```
64
65	Catch and continue if any GitHub call fails.
66
67	---
68
69	## Step 3: Score 10 dimensions
70
71	For each of the 10 dimensions in `references/scoring-rubric.md`, output `{score: 0–10, evidence: [...]}` where `evidence` is a list of 1–3 short strings citing what you found.
72
73	Example:
74	```json
75	"visual_proof": {
76	"score": 4,
77	"evidence": [
78	"Found 1 screenshot at line 89 (below the fold)",
79	"No animated demo / gif",
80	"No video"
81	]
82	}
83	```
84
85	Sum scores → total (0–100) → grade (A/B/C/D/F per rubric).
86
87	Save to `dx-roast/score.json` with this exact shape:
88
89	```json
90	{
91	"repo": "<owner/repo or local path>",
92	"scanned_at": "<ISO 8601 timestamp>",
93	"dimensions": {
94	"time_to_first_success": { "score": 7, "evidence": ["..."] },
95	"what_is_this_clarity": { "score": 5, "evidence": ["..."] },
96	"visual_proof": { "score": 4, "evidence": ["..."] },
97	"install_simplicity": { "score": 9, "evidence": ["..."] },
98	"runnable_quick_start": { "score": 8, "evidence": ["..."] },
99	"documentation_depth": { "score": 6, "evidence": ["..."] },
100	"examples": { "score": 5, "evidence": ["..."] },
101	"community_signals": { "score": 3, "evidence": ["..."] },
102	"trust_signals": { "score": 7, "evidence": ["..."] },
103	"marketing_signals"

People who liked this also viewed

6 picks

Type

dx-roaster

13installs59forks

View on GitHub

How to install dx-roaster?

varnan-tech/opendirectory/dx-roaster

$npx -y skills add Varnan-Tech/opendirectory --skill dx-roaster

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# dx-roaster
2
3	Roasts a GitHub repo's developer experience. Produces four files: `roast.md`, `action-plan.md`, `ideal-readme.md`, `score.json`. Built for OSS maintainers and dev-tool founders preparing for Show HN, Product Hunt, or any launch where the first five minutes matter.
4
5	---
6
7	## Critical rules (read before every run)
8
9	1. Read all three reference files BEFORE scoring. `references/scoring-rubric.md`, `references/roast-voice-guide.md`, `references/readme-template.md`. The rubric is non-negotiable — use the exact 10 dimensions.
10
11	2. Score every dimension on the 0–10 scale. Never skip. Missing data = score 0, not "N/A".
12
13	3. Roast must cite specific findings. Quote bad lines verbatim. Reference file paths. Cite line numbers when possible. Never write generic platitudes like "documentation could be better".
14
15	4. Tone matters. Default `honest`. Switch voice per `references/roast-voice-guide.md`. Brutal is funny but never personal. Kind is encouraging but still specific.
16
17	5. Action plan is impact-ranked, not severity-ranked. Use `impact × (1 / effort)`. A 5-minute fix that adds 10 points ranks above a 2-hour fix that adds 12 points.
18
19	6. Never invent URLs. When writing `ideal-readme.md`, use `<placeholder>` syntax for missing links. Roast will tell maintainer to fill in.
20
21	7. Never name maintainers or contributors. Roast the work, not the people.
22
23	8. Output exactly four files in `dx-roast/`. Do not write to other locations. Do not skip files.
24
25	---
26
27	## Step 1: Intake
28
29	Required:
30	- `repo` — local path (default `.`) OR `https://github.com/owner/repo` URL
31
32	Optional:
33
34	\| Parameter \| Default \| Description \|
35	\|---\|---\|---\|
36	\| `tone` \| `honest` \| `brutal` / `honest` / `kind` \|
37	\| `output_dir` \| `dx-roast/` \| Where to write the 4 output files \|
38	\| `compare_to` \| none \| Optional second repo URL for benchmark comparison \|
39
40	If `repo` is a URL: clone to a temp dir using `git clone <url> /tmp/dx-roast-<short-hash>`. Scan from there. Delete temp dir after run.
41
42	If `repo` is `.` or local path: scan in place, do not modify anything.
43
44	---
45
46	## Step 2: Discover
47
48	Read these files (each may be absent — note absence):
49	- `README.md` — full contents, structure (headings), code blocks, image refs, badge URLs
50	- `CONTRIBUTING.md`, `CODE_OF_CONDUCT.md`
51	- `.github/ISSUE_TEMPLATE/`, `.github/PULL_REQUEST_TEMPLATE.md`, `.github/CODEOWNERS`
52	- `package.json` (description, homepage, repository, license, scripts), `pyproject.toml`, `Cargo.toml`, `go.mod`
53	- `LICENSE` (existence + type)
54	- `docs/` folder contents (just file names + sizes)
55	- `examples/` or `samples/` or `cookbook/` folder
56	- `.github/workflows/*.yml` (just names + brief CI presence check)
57
58	If `repo` is a URL, also fetch via `gh`:
59	```bash
60	gh api repos/<owner>/<repo> --jq '{stars: .stargazers_count, forks: .forks_count, watchers: .subscribers_count, pushed_at, description, homepage, topics, default_branch}'
61	gh api repos/<owner>/<repo>/commits --jq '.[0].commit.author.date' \| head -1
62	gh api repos/<owner>/<repo>/contributors --paginate --jq '[.[] \| .login] \| length'
63	```
64
65	Catch and continue if any GitHub call fails.
66
67	---
68
69	## Step 3: Score 10 dimensions
70
71	For each of the 10 dimensions in `references/scoring-rubric.md`, output `{score: 0–10, evidence: [...]}` where `evidence` is a list of 1–3 short strings citing what you found.
72
73	Example:
74	```json
75	"visual_proof": {
76	"score": 4,
77	"evidence": [
78	"Found 1 screenshot at line 89 (below the fold)",
79	"No animated demo / gif",
80	"No video"
81	]
82	}
83	```
84
85	Sum scores → total (0–100) → grade (A/B/C/D/F per rubric).
86
87	Save to `dx-roast/score.json` with this exact shape:
88
89	```json
90	{
91	"repo": "<owner/repo or local path>",
92	"scanned_at": "<ISO 8601 timestamp>",
93	"dimensions": {
94	"time_to_first_success": { "score": 7, "evidence": ["..."] },
95	"what_is_this_clarity": { "score": 5, "evidence": ["..."] },
96	"visual_proof": { "score": 4, "evidence": ["..."] },
97	"install_simplicity": { "score": 9, "evidence": ["..."] },
98	"runnable_quick_start": { "score": 8, "evidence": ["..."] },
99	"documentation_depth": { "score": 6, "evidence": ["..."] },
100	"examples": { "score": 5, "evidence": ["..."] },
101	"community_signals": { "score": 3, "evidence": ["..."] },
102	"trust_signals": { "score": 7, "evidence": ["..."] },
103	"marketing_signals"

People who liked this also viewed

6 picks

Type