task-forest
Maintains a repo-local task forest or task DAG for the current workspace
Install / Use
npx skills add dongshuyan/compass-skills --skill task-forestInstalls into whichever agent you are using.
SKILL.md
Installable skill definition
Quality Score
Category
Development & EngineeringSupported Platforms
Our assessment of task-forest
task-forest scores 88/100 on our quality scale, 2005th of 4,702 Development & Engineering skills we index (top 43%).
Its SKILL.md is 6.3 KB long, well organised into 8 sections with 4 code examples: a thorough specification that gives an agent plenty to work with.
It has 739 GitHub stars, a meaningful sign that others use it.
Maintenance, license and trust
- The repository was last updated about 3 months ago, so task-forest is actively maintained.
- Our last check on 2026-11-01 found the source still online.
- It is released under the MIT license, a permissive license that allows use, modification and commercial use with attribution.
- Its trust signals score 100/100, with no cautions. These come from repository metadata, not a code audit — read the skill file before letting an agent act on it.
Safety scan
No issues foundOur scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful.
AI review by kimi-k2.7-code on 2026-10-07. Automated pattern scan on 2026-10-07. It catches known dangerous patterns, not every risk — read a skill before letting an agent act on it.
task-forest compared with similar skills
All 4 of these similar skills score higher than task-forest; compare them before choosing.
| Skill | Score | Stars | Updated | Format |
|---|---|---|---|---|
| task-forest (this skill)by dongshuyan | 88 | 739 | 3mo ago | SKILL.md |
| taste-skillby Leonxlnx | 100 | 117.6k | 9d ago | CLAUDE.md |
| ai-job-searchby ai-job-search-dev | 100 | 72.3k | today | CLAUDE.md |
| claude-howtoby luongnv89 | 100 | 47.4k | 3d ago | CLAUDE.md |
| lobehubby lobehub | 100 | 85.7k | today | MCP Server |
Frequently asked questions
- How do I install task-forest?
- Run
npx skills add dongshuyan/compass-skills --skill task-forest. The install tabs above show the steps for each supported agent. - Which AI agents does task-forest work with?
- It is written for Universal, as a SKILL.md file. Other agents that read the same format can often use it too.
- Is task-forest safe to use?
- Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful. It is MIT-licensed and scores 100/100 on trust signals. Skills are instructions an agent will follow, so read the file before installing it and do not approve commands you do not understand.
- Is task-forest still maintained?
- The repository was last updated about 3 months ago, so task-forest is actively maintained.
Skill content
View source on GitHubname: task-forest description: Maintains a repo-local task forest or task DAG for the current workspace. Use when the user asks to initialize or update a task forest, close a session, summarize evolving project work, align a request with a global goal, track progress/history/deviations/todos, save or apply a task proposal, or export the client-readable task-forest HTML. Do not use for executing the tracked tasks themselves or for generic HTML work.
Task Forest
Purpose
Maintain the current workspace's task structure, proposals, history, and progress. Produce one standalone HTML deliverable that a first-time reader can understand without opening another task list or internal file.
Use the user's language for task content, proposals, reports, and HTML. Default to Chinese when unknown.
Portability
This skill is agent-agnostic. Install the whole task-forest directory in any host-supported skill location, then resolve scripts, references, and assets from the directory containing this SKILL.md. Do not assume a specific agent name, skill root, home-directory layout, shell, path separator, or operating system.
Use an available Python 3 launcher on the host (python3, python, or py -3). The scripts use the Python standard library and support macOS, Linux, and Windows. To label task history with the calling agent, set COMPASS_AGENT_NAME or AGENT_NAME, or pass --actor; otherwise the neutral value agent is used.
Core Rules
- Read and write task-forest data only through
scripts/task_forest.py; never hand-edit.agent-workbench/task-forest/canonical files. - Use one primary
child_ofparent per node,contributes_tofor secondary ownership, anddepends_onfor prerequisites. - Save graph changes as a proposal and wait for user confirmation before
proposal-apply --yes. - Keep low-confidence inference in a question or proposal. Record material execution drift as a deviation.
- Write visible task titles and purposes in plain language. A reader must understand what the task delivers, why it exists, and what has been completed without knowing internal codes such as
P04or reading another file. - Export one HTML surface:
exports/task-forest.html. It showsdone,in_progress, and their necessarychild_ofancestors, with history playback. Do not expose internal discussions, evidence, queues, filesystem paths, sessions, or proposal content in the HTML. - Keep HTML interactions read-only. Formal changes always return through the proposal workflow.
- Keep task data and discovery metadata repo-local by default. Cross-workspace discovery is optional: enable it only after the user explicitly opts in by setting
TASK_FOREST_ENABLE_GLOBAL_REGISTRY=1. This writes lightweight workspace paths and health summaries toAGENT_WORKBENCH_DB, or to~/.agent-workbench/agent-workbench.sqlite3when that variable is unset; it never stores task content.
Main Workflow
When initializing, updating, or closing a session:
- Run
init. - Read
list --jsonandtodo --json. - Identify the global goal served by the session and the task structure that must remain visible.
- When the workspace has an authoritative task list, preserve its meaningful
goal -> phase -> module -> concrete taskhierarchy and sibling order for everydoneorin_progresstask. Include necessary ancestors, omit wholly unstarted branches, and attach extra fixes under the feature they improve. Never collapse several phases into one node or rely on edge creation order. If one sibling needsdisplay_order, set a unique numeric value for the whole sibling group; partial, duplicate, or invalid values must fail validation. - Make every visible node independently understandable. Use a clear title plus
summaryorpurpose; add outcomes or acceptance criteria when they clarify delivery. Treat internal codes as secondary labels, not as the task name. - Show and save a proposal. Do not apply it before confirmation.
- After confirmation, run
proposal-apply --yes,validate, andexport. - Return the proposal path and the single HTML path.
Use $task-clarifier when user intent or the target global goal is genuinely unclear.
Commands
Resolve <skill-dir> from this file and use an available Python 3 executable. The examples use python3; substitute the host's available launcher when needed.
python3 <skill-dir>/scripts/task_forest.py init
python3 <skill-dir>/scripts/task_forest.py list --json
python3 <skill-dir>/scripts/task_forest.py todo --json
python3 <skill-dir>/scripts/task_forest.py proposal-save --proposal-file /path/to/proposal.json
python3 <skill-dir>/scripts/task_forest.py proposal-apply <proposal-id> --yes
python3 <skill-dir>/scripts/task_forest.py validate
python3 <skill-dir>/scripts/task_forest.py export
The default workspace is the current directory. Use --workspace only when another workspace is explicit. Use --root only when the caller explicitly selected a non-default task-forest root.
Global registry integration is off by default. TASK_FOREST_DISABLE_GLOBAL_REGISTRY=1 remains an explicit override when a host sets the enable flag globally.
Outputs
The user-facing artifacts are:
proposals/<proposal_id>.json
exports/task-forest.html
The exporter also maintains three internal compatibility files for gap-router and local-agent-control-room:
exports/task-forest.graph.json
exports/task-forest.todos.json
exports/task-forest.timeline.json
Do not present those JSON files as delivery artifacts unless the user explicitly asks for machine-readable data.
References and Validation
- Read
references/schema.mdfor node fields, proposal actions, and canonical invariants. - Read
references/goal-alignment.mdonly when judging global-goal fit or competing candidate plans. - Read
references/node-types.mdonly when node classification is unclear. - Read
references/concurrency.mdbefore resolving stale proposals or concurrent writes. - Read
references/html-visualization-contract.mdwhen changing or validating the HTML. - Read
references/integration-contract.mdonly when changing JSON or registry compatibility.
For HTML or exporter changes, run:
python3 <skill-dir>/scripts/validate_task_forest_export.py --skill-dir <skill-dir>
The HTML remains a derived, read-only view. Canonical task data and proposal history stay repo-local.
Related Skills
taste-skill
117.6kAgent skills that give coding agents taste: art direction, reference images and a browser check instead of generic AI pages
ai-job-search
72.3kThe job search that runs on your machine. AI job application framework built on Claude Code: evaluate postings, tailor CVs, write cover letters, prep interviews. Fork it and own it.
claude-howto
47.4kA visual, example-driven guide to Claude Code — from basic concepts to advanced agents, with copy-paste templates that bring immediate value.
lobehub
85.7k🤯 LobeHub - Super-Agents on Your Team. Add super-agents to the chats your team already uses.They take on work by themselves and get better every week.
Languages
Trust signals
From repository metadata: license, adoption, age and documentation. Not a code audit — see the Safety scan above for what the skill file itself contains.
