planning

Name: planning
Author: langchain-ai

27kstars3.7kforks

How to install planning?

langchain-ai/deepagents/planning

$npx -y skills add langchain-ai/deepagents --skill planning

Installs into the current project.

Summary

Break down a coding task into a structured implementation plan with clear steps, file identification, and risk assessment.

Files · 1

SKILL.md

1	# Planning Skill
2
3	Use this skill when starting a new coding task to create a thorough implementation plan.
4
5	## Steps
6
7	### 1. Understand the Task
8	- Read the issue/task description completely
9	- Identify the expected outcome and acceptance criteria
10	- Note any constraints or requirements mentioned
11
12	### 2. Explore the Codebase
13	- Find the repository root and read the project structure
14	- Identify the tech stack (language, framework, test runner)
15	- Read README, CONTRIBUTING, or similar docs if they exist
16	- Find existing tests to understand testing patterns
17
18	### 3. Identify Relevant Files
19	- Use `grep` to find code related to the task
20	- Read the most relevant files (entry points, related modules)
21	- Identify which files need to be modified vs. created
22	- Check for existing patterns you should follow
23
24	### 4. Write the Plan
25	Use `write_todos` to create a structured plan:
26
27	```
28	write_todos([
29	"1. <specific change in specific file>",
30	"2. <next specific change>",
31	"3. Write tests for <feature>",
32	"4. Run test suite and fix failures",
33	"5. Review all changes"
34	])
35	```
36
37	### 5. Assess Risks
38	- Are there breaking changes?
39	- Are there edge cases to handle?
40	- Does this affect other parts of the codebase?
41	- Flag anything uncertain for review
42
43	## Guidelines
44
45	- Plans should have 3-10 concrete steps
46	- Each step should be specific enough to execute without further planning
47	- Include test writing and test running as explicit steps
48	- End with a review/verification step

6 picks

Type