cm-skill-chain

Name: cm-skill-chain
Author: tody-agent

47stars23forks

How to install cm-skill-chain?

tody-agent/codymaster/cm-skill-chain

$npx -y skills add tody-agent/codymaster --skill cm-skill-chain

Installs into the current project.

Summary

Skill Chain Engine — compose skills into automated pipelines. One task triggers multi-skill workflows with progress tracking, auto-detection, and step management.

Files · 1

View on GitHub

SKILL.md

1	# Skill Chain Engine
2
3	> TRIZ #40 Composite Materials — Skills compose into pipelines.
4	> One command → full workflow → automated step progression.
5
6	## When to Use
7
8	ALWAYS trigger for: chain, pipeline, workflow, multi-step, end-to-end,
9	"run the whole thing", "full process", "feature pipeline", "bug fix flow",
10	"from scratch to deploy", "brainstorm to ship", "skill chain", "full pipeline"
11
12	## Quick Reference
13
14	\| Command \| Description \|
15	\|---------\|-------------\|
16	\| `cody chain list` \| Show all available chains \|
17	\| `cody chain info <id>` \| Show chain pipeline details \|
18	\| `cody chain auto "task"` \| Auto-detect best chain & start \|
19	\| `cody chain start <id> "task"` \| Start specific chain \|
20	\| `cody chain status [exec-id]` \| Show progress \|
21	\| `cody chain advance <exec-id>` \| Complete current step, move to next \|
22	\| `cody chain skip <exec-id>` \| Skip current step \|
23	\| `cody chain abort <exec-id>` \| Cancel chain \|
24	\| `cody chain history` \| View past chain runs \|
25
26	## Naming Convention
27
28	Step names use short-form identifiers — the `cm-` prefix is stripped by convention.
29	Full skill names: `cm-brainstorm-idea`, `cm-planning`, `cm-tdd`, `cm-execution`, `cm-quality-gate`,
30	`cm-safe-deploy`, `cm-debugging`, `cm-content-factory`, `cm-ads-tracker`, `cm-project-bootstrap`,
31	`cm-code-review`. All skills now use the `cm-` prefix convention.
32
33	## Built-in Chains
34
35	### 🚀 feature-development (up to 3 active steps)
36	`brainstorm-idea* → planning → tdd → execution → quality-gate → safe-deploy*`
37	*optional steps — only activated when task context scores them relevant
38
39	### 🐛 bug-fix (3 steps)
40	`debugging → tdd → quality-gate`
41
42	### 📝 content-launch (3 steps)
43	`content-factory → ads-tracker → cm-cro-methodology`
44
45	### 🏗️ new-project (up to 3 active steps)
46	`project-bootstrap → planning → tdd → execution → quality-gate → safe-deploy*`
47	*optional steps selected by task relevance
48
49	### 🔍 cm-code-review (3 steps)
50	`cm-code-review → quality-gate → safe-deploy`
51
52	## Intelligent Skill Selection (v5.1)
53
54	Chains no longer execute every step blindly. `selectTopSkills()` dynamically selects the top 3 most relevant steps for each task:
55
56	```
57	Task: "fix login timeout bug"
58	→ Scores each step by keyword overlap with task description
59	→ Mandatory steps (condition='always', optional=false) always included first
60	→ Optional steps ranked by relevance score, capped at 3 total
61	→ Result: debugging (score 105) → tdd (score 101) → quality-gate (score 100)
62	```
63
64	Why it matters (SkillsBench research):
65	- 2-3 focused skills → +18.6pp task performance
66	- 4+ skills → +5.9pp
67	- Monolithic loading → -2.9pp
68
69	If a chain has more than 3 mandatory steps, all mandatory steps run and a performance advisory is logged.
70
71	## Workflow
72
73	1. Start: Use `chain auto` for auto-detection or `chain start` for specific chains
74	2. Execute: Work through each skill step, using `@[/skill-name]` to invoke
75	3. Advance: When step is done, run `chain advance <id> "summary"`
76	4. Repeat: Continue until all steps complete
77	5. Track: Use `chain status` to monitor progress anytime
78
79	## Integration with Other Skills
80
81	- cm-continuity: Chain progress persists across sessions via CONTINUITY.md working memory
82	- cm-execution: Each chain step delegates to cm-execution for actual implementation
83	- cm-quality-gate: Automatically runs at end of each development chain
84	- Context Bus (v5): Every chain automatically maintains `.cm/context-bus.json` — shared state across all steps
85
86	## Context Bus — Inter-Skill Coordination (v5)
87
88	When `chain start` runs, the context bus is initialized automatically:
89
90	```
91	chain start feature-development "add payment flow"
92	→ Creates .cm/context-bus.json with:
93	pipeline: "feature-development"
94	session_id: "<uuid>"
95	current_step: "brainstorm-idea"
96	shared_context: {}
97	resource_state: { skeleton_generated: null, learnings_indexed: null, ... }
98	```
99
100	When `chain advance` runs after each skill completes:
101
102	```
103	chain advance <exec-id> "summary of what was done"
104	→ Updates context-bus.json:
105	current_step: "planning" ← moved forward
106	shared_context.brainstorm-idea: { summary, affected_files, output_path }
107	```
108
109	What downstream skills gain:
110	- `cm-planning` can read brainstorm output path → no re-read of full filesystem
111	- `cm-tdd` can see which files planning created → targeted test generation
112	- `cm-quality-gate` knows exactly which files changed → focused review
113
114	Reading the bus:
115	```bash
116	cm continuity bus # terminal pretty-print
117	cm_bus_read # MCP tool (Codex Desktop)
118	cm://pipeline/current # URI resolver (in skill prompts)
119	```
120
121	Publishing to the bus (inside a skill):
122	```bash
123	cm_bus_write skill=cm-planning summary="tasks.md created" output_path=openspec/...
124	```
125
126	## For AI Agents
127
128	When dispatching task

People who liked this also viewed

6 picks

Type

lark-drive飞书云空间（云盘/云存储）：管理 Drive 文件和文件夹，包含上传/下载、创建文件夹、复制/移动/删除、查看元数据、评论/权限/订阅、标题、版本、飞书文档密级标签（secure labels）和本地文件导入。用户需要整理云盘目录、处理云空间资源 URL/token、判断链接类型/真实 token/标题，或导入 Word/Markdown/Excel/CSV/PPTX/.base 为 docx/sSkillsJul 2026382k16k
lark-sharedUse for lark-cli setup/auth tasks: auth login/status/logout, user vs bot identity, business-domain permissions (--domain, including all/docs/drive), missing scopes, revoking authorization, or handlingSkillsJul 2026382k16k
lark-whiteboard飞书画板：查询和编辑飞书云文档中的画板。支持导出画板为预览图片、导出原始节点结构、使用多种格式更新画板内容。当用户需要查看画板内容、导出画板图片、编辑画板时使用此 skill。不负责：飞书云文档内容编辑（lark-doc）、文档内嵌电子表格/Base（lark-sheets / lark-base）。SkillsJul 2026380k16k
lark-task飞书任务：管理任务、清单和任务智能体。创建待办任务、查看和更新任务状态、拆分子任务、组织任务清单、分配协作成员、上传任务附件、注册或注销任务智能体、更新任务智能体的主页数据、写入智能体任务记录。当用户需要创建待办事项、查看任务列表、跟踪任务进度、管理项目清单或给他人分配任务、为任务上传附件文件、注册注销任务智能体、更新智能体主页数据、写入任务记录时使用。SkillsJul 2026380k16k
lark-attendance飞书考勤打卡：查询自己的考勤打卡记录SkillsJul 2026330k16k
lark-apps妙搭（Spark/Miaoda）应用开发与托管：应用创建、本地全栈开发、云端生成迭代、创意设计（UI mockup / 可交互原型 / 线框图 / 落地页 / 仪表盘 / 幻灯片 deck / 视觉探索）、AI相关能力和飞书平台能力或者其他外部能力集成、日志/Trace/监控指标/PV/UV 查询、环境变量管理、应用角色与成员管理、自动化触发器（定时/记录变更/Webhook/飞书审批）。当用户要SkillsJul 2026227k16k

cm-skill-chain

47stars23forks

View on GitHub

How to install cm-skill-chain?

tody-agent/codymaster/cm-skill-chain

$npx -y skills add tody-agent/codymaster --skill cm-skill-chain

Installs into the current project.

Summary

Skill Chain Engine — compose skills into automated pipelines. One task triggers multi-skill workflows with progress tracking, auto-detection, and step management.

Files · 1

View on GitHub

SKILL.md

1	# Skill Chain Engine
2
3	> TRIZ #40 Composite Materials — Skills compose into pipelines.
4	> One command → full workflow → automated step progression.
5
6	## When to Use
7
8	ALWAYS trigger for: chain, pipeline, workflow, multi-step, end-to-end,
9	"run the whole thing", "full process", "feature pipeline", "bug fix flow",
10	"from scratch to deploy", "brainstorm to ship", "skill chain", "full pipeline"
11
12	## Quick Reference
13
14	\| Command \| Description \|
15	\|---------\|-------------\|
16	\| `cody chain list` \| Show all available chains \|
17	\| `cody chain info <id>` \| Show chain pipeline details \|
18	\| `cody chain auto "task"` \| Auto-detect best chain & start \|
19	\| `cody chain start <id> "task"` \| Start specific chain \|
20	\| `cody chain status [exec-id]` \| Show progress \|
21	\| `cody chain advance <exec-id>` \| Complete current step, move to next \|
22	\| `cody chain skip <exec-id>` \| Skip current step \|
23	\| `cody chain abort <exec-id>` \| Cancel chain \|
24	\| `cody chain history` \| View past chain runs \|
25
26	## Naming Convention
27
28	Step names use short-form identifiers — the `cm-` prefix is stripped by convention.
29	Full skill names: `cm-brainstorm-idea`, `cm-planning`, `cm-tdd`, `cm-execution`, `cm-quality-gate`,
30	`cm-safe-deploy`, `cm-debugging`, `cm-content-factory`, `cm-ads-tracker`, `cm-project-bootstrap`,
31	`cm-code-review`. All skills now use the `cm-` prefix convention.
32
33	## Built-in Chains
34
35	### 🚀 feature-development (up to 3 active steps)
36	`brainstorm-idea* → planning → tdd → execution → quality-gate → safe-deploy*`
37	*optional steps — only activated when task context scores them relevant
38
39	### 🐛 bug-fix (3 steps)
40	`debugging → tdd → quality-gate`
41
42	### 📝 content-launch (3 steps)
43	`content-factory → ads-tracker → cm-cro-methodology`
44
45	### 🏗️ new-project (up to 3 active steps)
46	`project-bootstrap → planning → tdd → execution → quality-gate → safe-deploy*`
47	*optional steps selected by task relevance
48
49	### 🔍 cm-code-review (3 steps)
50	`cm-code-review → quality-gate → safe-deploy`
51
52	## Intelligent Skill Selection (v5.1)
53
54	Chains no longer execute every step blindly. `selectTopSkills()` dynamically selects the top 3 most relevant steps for each task:
55
56	```
57	Task: "fix login timeout bug"
58	→ Scores each step by keyword overlap with task description
59	→ Mandatory steps (condition='always', optional=false) always included first
60	→ Optional steps ranked by relevance score, capped at 3 total
61	→ Result: debugging (score 105) → tdd (score 101) → quality-gate (score 100)
62	```
63
64	Why it matters (SkillsBench research):
65	- 2-3 focused skills → +18.6pp task performance
66	- 4+ skills → +5.9pp
67	- Monolithic loading → -2.9pp
68
69	If a chain has more than 3 mandatory steps, all mandatory steps run and a performance advisory is logged.
70
71	## Workflow
72
73	1. Start: Use `chain auto` for auto-detection or `chain start` for specific chains
74	2. Execute: Work through each skill step, using `@[/skill-name]` to invoke
75	3. Advance: When step is done, run `chain advance <id> "summary"`
76	4. Repeat: Continue until all steps complete
77	5. Track: Use `chain status` to monitor progress anytime
78
79	## Integration with Other Skills
80
81	- cm-continuity: Chain progress persists across sessions via CONTINUITY.md working memory
82	- cm-execution: Each chain step delegates to cm-execution for actual implementation
83	- cm-quality-gate: Automatically runs at end of each development chain
84	- Context Bus (v5): Every chain automatically maintains `.cm/context-bus.json` — shared state across all steps
85
86	## Context Bus — Inter-Skill Coordination (v5)
87
88	When `chain start` runs, the context bus is initialized automatically:
89
90	```
91	chain start feature-development "add payment flow"
92	→ Creates .cm/context-bus.json with:
93	pipeline: "feature-development"
94	session_id: "<uuid>"
95	current_step: "brainstorm-idea"
96	shared_context: {}
97	resource_state: { skeleton_generated: null, learnings_indexed: null, ... }
98	```
99
100	When `chain advance` runs after each skill completes:
101
102	```
103	chain advance <exec-id> "summary of what was done"
104	→ Updates context-bus.json:
105	current_step: "planning" ← moved forward
106	shared_context.brainstorm-idea: { summary, affected_files, output_path }
107	```
108
109	What downstream skills gain:
110	- `cm-planning` can read brainstorm output path → no re-read of full filesystem
111	- `cm-tdd` can see which files planning created → targeted test generation
112	- `cm-quality-gate` knows exactly which files changed → focused review
113
114	Reading the bus:
115	```bash
116	cm continuity bus # terminal pretty-print
117	cm_bus_read # MCP tool (Codex Desktop)
118	cm://pipeline/current # URI resolver (in skill prompts)
119	```
120
121	Publishing to the bus (inside a skill):
122	```bash
123	cm_bus_write skill=cm-planning summary="tasks.md created" output_path=openspec/...
124	```
125
126	## For AI Agents
127
128	When dispatching task

People who liked this also viewed

6 picks

Type