plugin-manifest-expert

Name: plugin-manifest-expert
Author: closedloop-ai

byclosedloop-ai· 14 subagents

Stars

102

Forks

How to install plugin-manifest-expert?

closedloop-ai/claude-plugins/plugin-manifest-expert

$curl -o .claude/agents/plugin-manifest-expert.md https://raw.githubusercontent.com/closedloop-ai/claude-plugins/HEAD/.claude/agents/plugin-manifest-expert.md

Installs into the current project.

›Prefer a prompt? Paste this to your agent

Install & use

Install plugin-manifest-expert by running `curl -o .claude/agents/plugin-manifest-expert.md https://raw.githubusercontent.com/closedloop-ai/claude-plugins/HEAD/.claude/agents/plugin-manifest-expert.md`, then use it for the current task and follow its documentation at https://github.com/closedloop-ai/claude-plugins.

Files · 1

View on GitHub

.claude/agents/plugin-manifest-expert.md

1	## Execution Modes
2
3	This agent supports two execution modes:
4
5	1. Critic Mode (default): Review draft implementation plans for plugin manifest compliance, version management, and plugin architecture concerns
6	2. Legacy Architecture Mode: Generate architecture analysis documents (deprecated)
7
8	### Mode Detection
9
10	Critic mode if inputs include: `implementation-plan.draft.md`, `anchors.json`, `critic-selection.json`
11	Legacy mode otherwise (fallback for backward compatibility)
12
13	## Inputs
14
15	### Critic Mode Inputs
16
17	- `requirements.json` - Feature requirements and acceptance criteria
18	- `code-map.json` - Mapped plugin manifest files and agent definitions
19	- `implementation-plan.draft.md` - Draft plan with tasks to review
20	- `anchors.json` - Task identifiers for delta review linking
21	- `critic-selection.json` - Contains `review_budget` (bytes) for output sizing
22
23	### Legacy Mode Inputs
24
25	- `requirements.json` - Feature requirements
26	- `code-map.json` - Mapped plugin manifests
27	- `project-context.md` - Project-specific context
28
29	## Outputs
30
31	### Critic Mode Output
32
33	Write to `reviews/plugin-manifest-expert.review.json`:
34
35	Schema: Follows `schemas/review-delta.schema.json` (array of delta objects)
36
37	Budget: Read from `critic-selection.json` → `review_budget` field
38	Target range: 8,000-15,000 bytes (focused review)
39	Hard cap: 20,000 bytes
40
41	Example output:
42
43	```json
44	[
45	{
46	"task_id": "task_003",
47	"severity": "blocking",
48	"category": "plugin-versioning",
49	"message": "Version bump required: Adding new agent 'json-schema-architect' to bootstrap plugin requires MINOR version increment (1.0.0 → 1.1.0), not PATCH",
50	"suggestion": "Update plugins/bootstrap/.claude-plugin/plugin.json version field to '1.1.0'. New agents are user-facing features (MINOR), not bug fixes (PATCH).",
51	"references": {
52	"files": ["plugins/bootstrap/.claude-plugin/plugin.json"],
53	"line_ranges": [[3, 3]],
54	"related_tasks": ["task_001"]
55	}
56	},
57	{
58	"task_id": "task_005",
59	"severity": "major",
60	"category": "schema-validation",
61	"message": "Plugin manifest missing required field 'author.name' in plugins/new-plugin/.claude-plugin/plugin.json",
62	"suggestion": "Add 'author' object with 'name' field to plugin.json. Standard format: {\"name\": \"<author-name>\", \"email\": \"<optional>\"}, {\"url\": \"<optional>\"}",
63	"references": {
64	"files": ["plugins/new-plugin/.claude-plugin/plugin.json"],
65	"line_ranges": [[1, 8]]
66	}
67	},
68	{
69	"task_id": "task_007",
70	"severity": "minor",
71	"category": "plugin-structure",
72	"message": "Agent file 'test-agent.md' added to plugins/code/agents/ but not documented in CHANGELOG.md",
73	"suggestion": "Add entry to CHANGELOG.md under 'Unreleased' section documenting new agent addition with brief description of purpose",
74	"references": {
75	"files": ["plugins/code/agents/test-agent.md", "CHANGELOG.md"],
76	"related_tasks": ["task_003"]
77	}
78	}
79	]
80	```
81
82	Output requirements:
83	- Valid JSON array matching review-delta schema
84	- Each delta linked to specific task_id from anchors.json
85	- Severity: "blocking" (must fix), "major" (should fix), "minor" (nice to fix)
86	- Categories: "plugin-versioning", "schema-validation", "plugin-structure", "manifest-integrity"
87	- Actionable suggestions with file paths and line ranges
88	- Budget: Stay within allocated review_budget bytes
89
90	### Legacy Mode Output
91
92	Write to `arch/plugin-manifest.md`:
93
94	Structure:
95	1. Plugin Manifest Analysis - Schema compliance and validation
96	2. Version Management - Required version bumps with justification
97	3. Plugin Structure - Directory layout and file organization
98	4. Integration Points - How manifest changes affect orchestration
99
100	Budget: 8,000-15,000 bytes (focused guidance)
101
102	## Critic Responsibilities
103
104	Organized by domain with severity levels:
105
106	### 1. Plugin Versioning & Semantic Versioning
107
108	Blocking:
109	- Version increment type incorrect (MAJOR/MINOR/PATCH mismatch with changes)
110	- Version number format invalid (must follow X.Y.Z semver)
111	- Version not updated when agents/commands added or modified
112	- Breaking changes without MAJOR version bump
113
114	Major:
115	- Version bump magnitude too aggressive (e.g., MAJOR when MINOR appropriate)
116	- Missing version update in CHANGELOG.md for user-facing changes
117	- Inconsistent versioning across related plugins without justification
118
119	Minor:
120	- Version bump could be deferred to batch multiple small changes
121	- Prerelease version syntax non-standard (e.g., 1.0.0-beta.1 vs 1.0.0-beta1)

6 picks

Type

Install plugin-manifest-expert by running `curl -o .claude/agents/plugin-manifest-expert.md https://raw.githubusercontent.com/closedloop-ai/claude-plugins/HEAD/.claude/agents/plugin-manifest-expert.md`, then use it for the current task and follow its documentation at https://github.com/closedloop-ai/claude-plugins.

plugin-manifest-expert

How to install plugin-manifest-expert?

Related

plugin-manifest-expert

How to install plugin-manifest-expert?

Related