Process Kit for GigaCode workflow
process-kit
A GigaCode workspace where the assistant takes a task in words, finds a ready process made of steps and runs it with a plan and a report in outputs
- 0 ★ on the repository
- Last commit
Extended access
Why this level
- The assistant runs processes and pre-commit and pre-push hooks in the working folder
- Jira and Bitbucket connections read task and repository data with a token from the GigaCode settings
Install
- Works in
- Services
In your terminal, with SkillFoxx CLI
npx skillfoxx add workflows/process-kitDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- The assistant runs processes and pre-commit and pre-push hooks in the working folder
- Jira and Bitbucket connections read task and repository data with a token from the GigaCode settings
This is third-party code. Review the repository files before installing.
What it does
The repository is opened in GigaCode as an ordinary project and the assistant works from the `.gigacode` directory: it holds the skills `pp-ai-init`, `pp-ai-run-process`, `pp-ai-create-process`, `pp-ai-create-action`, `pp-ai-create-adapter`, `pp-ai-mentor` and two pairs for promoting changes, plus pre-commit and `pre-push` hooks. A user writes a task in plain words such as "show the weekly report", the assistant picks a process from ready steps, shows a plan and waits for consent. Ready parts split into processes, actions and system descriptions. Your own processes go next to others with a `my-` prefix, the team part lives in the `team-<team>` branch and the shared part arrives by tag. Every run writes its result to a separate `outputs` folder with an `итог.md` file, and the run log sits in `.history`. Access to systems such as Jira and Bitbucket is added in GigaCode itself, and files keep only a reference to the secret. There is no mail connection and a letter is prepared as a draft. The layout is checked with `node tools/kit.js check`.
Who it is for. Teams in an internal environment with GigaCode who want to run repeating work processes, such as Jira reports, as a set of shared steps and their own additions.
Good fit when
- A weekly summary of overdue items in Jira is assembled and you want it with one phrase to the assistant
- A team wants a shared set of processes while each employee also has personal ones
- A new process or action is created with the `pp-ai-create-process` skill from a template
- A trial run goes on fictional data with no connection to systems
Not a fit when
- The work is not in GigaCode: the `.gigacode` directory is built for that agent
- Mail exchange is needed: there is no mail connection and a letter is only drafted
- The task is one-off and does not repeat, and setting up a process takes longer than the work itself
Example request
Show the report on overdue tasks of our team in Jira for the week.Limitations
The repository opens in GigaCode (CLI, VS Code or the app), Jira and Bitbucket need access added in the GigaCode settings, and service commands need Node.js. The assistant does not ask for tokens in chat and does not use one that was sent.
After access is added or a new file lands in `.gigacode/skills` the client has to be restarted, otherwise the command does not appear. In the README the clone address is given as a template, GigaCode versions are not named and no license is visible. The shared part updates by merging a tag with `git merge v0.2.0`, and conflicts are the user to resolve.
How to disable. Close the `process-kit` folder in GigaCode and delete it: the assistant does not change global computer settings.
FAQ
How do I start with Process Kit?
Clone the repository, open the folder in GigaCode as an ordinary project and write a task in words. The assistant leads the setup from there.
Where are results saved?
In the `outputs` folder: each run gets its own subfolder with a date and a task name, with an `итог.md` file inside. The log of all runs sits in `.history`.
How are Jira and Bitbucket tokens stored?
In the GigaCode settings. Files hold only a secret reference like `$secret:pp-ai-jira/x-jira-token`.
How do I add my own process?
Create it with the `pp-ai-create-process` skill. Name your files with the `my-` prefix so they do not conflict with shared ones on update.
How do I get shared part updates?
Run `git fetch --tags origin` and merge the needed tag, for example `git merge v0.2.0`, then run `node tools/kit.js check`.
Discussion
No questions or reviews yet.
Related
All GigaChat toolsGigaCode project flows with hooks, gates and OpenSpec
gigacodeAgents
A monorepo of flows for the GigaCode CLI: development with stages and hard gates, a work progress observer and reverse analysis of code into a spec
GigaCode Bridge MCP
Gigacode-bridge-mcp
MCP server that delegates isolated code edits to GigaCode CLI in temporary Git worktrees and returns a patch for independent review
QA Agent for the GigaCode CLI: 14 roles and 6 skills for testing
QaAgentGigacodeCli
Requirements-based testing in GigaCode: from analysis and autotest review to a conclusion, with duplicate search before a bug is filed in Jira
Gigacode Task Commit Check
gigaide-precommit-gigacode
GigaIDE and JetBrains plugin that runs chosen GigaCode CLI tasks before a commit with the change context passed as JSON
A single Node.js file turns the installed GigaCode CLI into a local OpenAI server with streaming and a shared dialogue history
Eight skills for coding against the GigaChat API: choosing a layer, access setup, SDK, LangChain, gpt2giga and Deep Agents
README in short
The README covers installation, the `outputs` folder structure, connecting Jira and Bitbucket, updating the shared part by tag and the team branch, the personal branch and limits, with a second part addressed to the kit developer.