Command Claude CodeOpenCode
Register a known code-agent model in models .json, optionally promote it to default across config/seed/DEFAULTAGENTSTRING.
an agentic IDE in your terminal, markdown tasks with YAML frontmatter, git-native, zero infrastructure. Works with Claude Code, OpenCode, and Codex CLI.
This repository also configures its own agents. See what aitasks tells them →
Command Claude CodeOpenCode
Register a known code-agent model in models .json, optionally promote it to default across config/seed/DEFAULTAGENTSTRING.
Command Claude CodeOpenCode
Audit and port aitask skill wrappers across code-agent trees, plus helper-script whitelist coverage.
Command Claude CodeOpenCode
Generate a changelog entry by analyzing commits and archived plans since the last release.
Command Claude CodeOpenCode
Contribute local aitasks framework changes back to the upstream repository by opening structured GitHub issues.
Command Claude CodeOpenCode
Analyze a contribution issue, find related issues, and import as grouped or single task.
Command Claude CodeOpenCode
Create a new AI task file with automatic numbering and proper metadata.
Command Claude CodeOpenCode
Analyze tasks landed since the last release and create one documentation task covering website docs gaps.
Command Claude CodeOpenCode
Explain files in the project: functionality, usage examples, and code evolution history traced through aitasks.
Command Claude CodeOpenCode
Explore the codebase interactively, then create a task for implementation.
Command Claude CodeOpenCode
Identify and merge related tasks into a single task, then optionally execute it.
Command Claude CodeOpenCode
Procedure-backed verifier for the docsupdated gate — inspects a task's change, updates the project's documentation per the configured doc-update spec (confirming with the user), and records the gate result. Run by the attended agent (task-workflow / aitask-resume), not the headless engine.
Command Claude CodeOpenCode
Scaffold for authoring a new gate verifier. Documents the orchestrator's verifier contract (positional args, exit codes, sidecar log, ledger append) and provides a copy-me script template.
Command Claude CodeOpenCode
Learn a new skill from sources — a local file, a URL, a repo file/dir, or a tmux pane (capturing the workflow an agent just ran) — and generate a complete static SKILL.md.
Command Claude CodeOpenCode
Select the next AI task for implementation from the aitasks/ directory.
Command Claude CodeOpenCode
Pick and implement a task in remote/non-interactive mode. All decisions from execution profile - no AskUserQuestion calls.
Command Claude CodeOpenCode
Pick and implement a task on Claude Code Web. Zero interactive prompts. No cross-branch operations — stores task data locally in .aitask-data-updated/.
Command Claude CodeOpenCode
Create an aitask from a pull request by analyzing PR data and generating a structured task with implementation plan.
Command Claude CodeOpenCode
Run QA analysis on any task — analyze changes, discover test gaps, run tests, and create follow-up test tasks.
Command Claude CodeOpenCode
Research latest AI code agent models via web and update models.json configuration files.
Command Claude CodeOpenCode
Resume an in-flight task from its gate-ledger checkpoint (programmatic re-entry surface for testing / TUI / board ops).
Command Claude CodeOpenCode
Revert changes associated with completed tasks — fully or partially.
Command Claude CodeOpenCode
Review code using configurable review guides, then create tasks from findings.
Command Claude CodeOpenCode
Classify a review guide file by assigning metadata and finding similar existing guides.
Command Claude CodeOpenCode
Import external content (file, URL, or repository directory) as a reviewguide with proper metadata.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: