review-skill

Name: review-skill
Author: mongodb

161stars29forks

How to install review-skill?

mongodb/agent-skills/review-skill

$npx -y skills add mongodb/agent-skills --skill review-skill

Installs into the current project.

Summary

Review a proposed Agent Skill for structural validity and content quality before publishing. Runs the skill-validator CLI to check for structural issues, scores the skill with an LLM judge, and interprets results to advise SMEs on what to address. Use when a user wants to review,

Files · 1

View on GitHub

SKILL.md

1	# Review Skill Workflow
2
3	You are helping an SME review an Agent Skill before publishing. This is a
4	multi-step process: determine environment, verify prerequisites, run structural
5	validation, review content, optionally run LLM scoring, and interpret results.
6	Follow every step in order.
7
8	## Step 0: Determine Environment
9
10	Check for saved configuration:
11
12	```bash
13	cat ~/.config/skill-validator/review-state.yaml 2>/dev/null
14	```
15
16	If the state file exists with `prereqs_passed: true`, offer:
17
18	> Found saved settings — configured for [full/structural-only] reviews.
19	>
20	> 1. Continue with saved settings — skip to Step 2
21	> 2. Re-run prerequisite checks
22	> 3. Change environment — switch between full and structural-only
23
24	Option 1: read `llm_scoring` from the file and skip to Step 2.
25	Options 2-3: continue below.
26
27	If no state file exists, or the user chose to re-check/change, ask:
28
29	> LLM scoring evaluates content quality across multiple dimensions.
30	>
31	> 1. Yes, run LLM scoring — full review with LLM scoring
32	> 2. No, skip LLM scoring — structural validation only
33
34	Option 1: set `LLM_SCORING=true`.
35	Option 2: set `LLM_SCORING=false`. Run Step 1a only, then jump to Step 2.
36
37	## Step 1: Verify Prerequisites
38
39	### 1a. Check for `skill-validator` binary
40
41	```bash
42	skill-validator --version
43	```
44
45	If not found, search common locations (`/usr/local/bin`, `/opt/homebrew/bin`,
46	`~/go/bin`). If found but not on PATH, tell the user. If not found anywhere,
47	follow [references/install-skill-validator.md](references/install-skill-validator.md).
48
49	If `--version` is not at least v1.5.1, help the user upgrade with
50	`brew upgrade skill-validator` or
51	`go install github.com/agent-ecosystem/skill-validator/cmd/skill-validator@latest`.
52
53	Do NOT proceed until this succeeds.
54
55	### 1b. Check for `claude` CLI (LLM scoring only)
56
57	If `LLM_SCORING=true`, verify the Claude CLI is available:
58
59	```bash
60	claude --version
61	```
62
63	If not found, tell the user to install Claude Code:
64
65	- macOS: `curl -fsSL https://claude.ai/install.sh \| bash`
66	- Other platforms: follow the [Claude Code quickstart guide](https://code.claude.com/docs/en/quickstart)
67
68	The user must authenticate by running `claude` interactively before continuing.
69
70	Do NOT proceed with LLM scoring until this succeeds.
71
72	### Save state after prerequisites pass
73
74	Persist state so future runs skip this step. Replace `<true or false>` with
75	the actual `LLM_SCORING` value:
76
77	```bash
78	mkdir -p ~/.config/skill-validator
79	cat > ~/.config/skill-validator/review-state.yaml << 'EOF'
80	prereqs_passed: true
81	llm_scoring: <true or false>
82	EOF
83	```
84
85	## Step 2: Locate the Skill
86
87	Ask the user for the path to the skill they want to review, unless they have
88	already provided it. Verify the path contains a `SKILL.md` file:
89
90	```bash
91	ls <path>/SKILL.md
92	```
93
94	If `SKILL.md` does not exist at the given path, tell the user this is not a
95	valid skill directory and ask them to provide the correct path.
96
97	## Step 3: Run Structural Validation
98
99	Run the full check suite:
100
101	```bash
102	skill-validator check <path>
103	```
104
105	Capture the exit code:
106
107	\| Exit code \| Meaning \|
108	\|-----------\|---------\|
109	\| 0 \| Clean — no errors or warnings \|
110	\| 1 \| Errors found — must fix before publishing \|
111	\| 2 \| Warnings only — review but not blocking \|
112	\| 3 \| CLI/usage error — check the command \|
113
114	Exit 0: proceed. Exit 2: note warnings, proceed. Exit 1: list errors — these
115	are blocking. The user must fix them before the skill can be published. Do NOT
116	proceed to LLM scoring if exit code is 1.
117
118	## Step 4: Content Review
119
120	Read the SKILL.md and any reference files, then evaluate each check below.
121	Report which checks pass and which do not, with specific details on what is
122	missing.
123
124	\| Check \| Criteria \|
125	\|-------\|----------\|
126	\| Examples \| Does the skill provide examples of expected inputs and outputs? \|
127	\| Edge cases \| Does the skill document common edge cases or failure modes? \|
128	\| Scope-gating \| Does the skill define when to stop/continue, prerequisites, and conditions for branching paths? \|
129	\| MongoDB data access \| If the skill needs MongoDB contextual data, does it instruct agents to use the MCP server for auth and tool calls? Skip if not applicable. \|
130
131	Flag any failing checks as areas the SME should address. These are not blocking
132	but should be resolved before publishing for best results.
133
134	## Step 5: LLM Scoring and Interpretation
135
136	If `LLM_SCORING=false`, skip to Step 6.
137
138	If `LLM_SCORING=true`, follow the "Run LLM Scoring" and "Int

People who liked this also viewed

6 picks

Type

review-skill

161stars29forks

View on GitHub

How to install review-skill?

mongodb/agent-skills/review-skill

$npx -y skills add mongodb/agent-skills --skill review-skill

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Review Skill Workflow
2
3	You are helping an SME review an Agent Skill before publishing. This is a
4	multi-step process: determine environment, verify prerequisites, run structural
5	validation, review content, optionally run LLM scoring, and interpret results.
6	Follow every step in order.
7
8	## Step 0: Determine Environment
9
10	Check for saved configuration:
11
12	```bash
13	cat ~/.config/skill-validator/review-state.yaml 2>/dev/null
14	```
15
16	If the state file exists with `prereqs_passed: true`, offer:
17
18	> Found saved settings — configured for [full/structural-only] reviews.
19	>
20	> 1. Continue with saved settings — skip to Step 2
21	> 2. Re-run prerequisite checks
22	> 3. Change environment — switch between full and structural-only
23
24	Option 1: read `llm_scoring` from the file and skip to Step 2.
25	Options 2-3: continue below.
26
27	If no state file exists, or the user chose to re-check/change, ask:
28
29	> LLM scoring evaluates content quality across multiple dimensions.
30	>
31	> 1. Yes, run LLM scoring — full review with LLM scoring
32	> 2. No, skip LLM scoring — structural validation only
33
34	Option 1: set `LLM_SCORING=true`.
35	Option 2: set `LLM_SCORING=false`. Run Step 1a only, then jump to Step 2.
36
37	## Step 1: Verify Prerequisites
38
39	### 1a. Check for `skill-validator` binary
40
41	```bash
42	skill-validator --version
43	```
44
45	If not found, search common locations (`/usr/local/bin`, `/opt/homebrew/bin`,
46	`~/go/bin`). If found but not on PATH, tell the user. If not found anywhere,
47	follow [references/install-skill-validator.md](references/install-skill-validator.md).
48
49	If `--version` is not at least v1.5.1, help the user upgrade with
50	`brew upgrade skill-validator` or
51	`go install github.com/agent-ecosystem/skill-validator/cmd/skill-validator@latest`.
52
53	Do NOT proceed until this succeeds.
54
55	### 1b. Check for `claude` CLI (LLM scoring only)
56
57	If `LLM_SCORING=true`, verify the Claude CLI is available:
58
59	```bash
60	claude --version
61	```
62
63	If not found, tell the user to install Claude Code:
64
65	- macOS: `curl -fsSL https://claude.ai/install.sh \| bash`
66	- Other platforms: follow the [Claude Code quickstart guide](https://code.claude.com/docs/en/quickstart)
67
68	The user must authenticate by running `claude` interactively before continuing.
69
70	Do NOT proceed with LLM scoring until this succeeds.
71
72	### Save state after prerequisites pass
73
74	Persist state so future runs skip this step. Replace `<true or false>` with
75	the actual `LLM_SCORING` value:
76
77	```bash
78	mkdir -p ~/.config/skill-validator
79	cat > ~/.config/skill-validator/review-state.yaml << 'EOF'
80	prereqs_passed: true
81	llm_scoring: <true or false>
82	EOF
83	```
84
85	## Step 2: Locate the Skill
86
87	Ask the user for the path to the skill they want to review, unless they have
88	already provided it. Verify the path contains a `SKILL.md` file:
89
90	```bash
91	ls <path>/SKILL.md
92	```
93
94	If `SKILL.md` does not exist at the given path, tell the user this is not a
95	valid skill directory and ask them to provide the correct path.
96
97	## Step 3: Run Structural Validation
98
99	Run the full check suite:
100
101	```bash
102	skill-validator check <path>
103	```
104
105	Capture the exit code:
106
107	\| Exit code \| Meaning \|
108	\|-----------\|---------\|
109	\| 0 \| Clean — no errors or warnings \|
110	\| 1 \| Errors found — must fix before publishing \|
111	\| 2 \| Warnings only — review but not blocking \|
112	\| 3 \| CLI/usage error — check the command \|
113
114	Exit 0: proceed. Exit 2: note warnings, proceed. Exit 1: list errors — these
115	are blocking. The user must fix them before the skill can be published. Do NOT
116	proceed to LLM scoring if exit code is 1.
117
118	## Step 4: Content Review
119
120	Read the SKILL.md and any reference files, then evaluate each check below.
121	Report which checks pass and which do not, with specific details on what is
122	missing.
123
124	\| Check \| Criteria \|
125	\|-------\|----------\|
126	\| Examples \| Does the skill provide examples of expected inputs and outputs? \|
127	\| Edge cases \| Does the skill document common edge cases or failure modes? \|
128	\| Scope-gating \| Does the skill define when to stop/continue, prerequisites, and conditions for branching paths? \|
129	\| MongoDB data access \| If the skill needs MongoDB contextual data, does it instruct agents to use the MCP server for auth and tool calls? Skip if not applicable. \|
130
131	Flag any failing checks as areas the SME should address. These are not blocking
132	but should be resolved before publishing for best results.
133
134	## Step 5: LLM Scoring and Interpretation
135
136	If `LLM_SCORING=false`, skip to Step 6.
137
138	If `LLM_SCORING=true`, follow the "Run LLM Scoring" and "Int

People who liked this also viewed

6 picks

Type