Agent skill

Start Jumbo Goal

by jumbocontext in jumbocontext/cli

A skill your agent uses when starting a Jumbo goal. An agent skill from jumbocontext/cli.

AGPL-3.0Auto-check passedDevelopment

Install Start Jumbo Goal

skills CLI
$ npx skills add jumbocontext/cli --skill start-jumbo-goal -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install jumbocontext/cli start-jumbo-goal --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/jumbocontext/cli.git skills-src && mkdir -p .claude/skills && cp -r skills-src/assets/skills/start-jumbo-goal .claude/skills/start-jumbo-goal && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
start-jumbo-goal
GitHub stars
276
Token cost
~1.2k tokens
SKILL.md length
520 words
Files
1
Skills in repo
13
Repo updated
First seen
Licence
AGPL-3.0

At a glance

A skill your agent uses when starting a Jumbo goal. An agent skill from jumbocontext/cli.

  • Works in 5 steps: Start the Goal → Understand the Implementation Instructions → Execute Within Scope → …
  • Starting a Jumbo goal
  • SKILL.md covers Why Structured Implementation…, Protocol and Rules
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Start Jumbo Goal is an agent skill from jumbocontext/cli. Use when starting a Jumbo goal. Loads goal context, scope, and architectural constraints, then guides the agent through disciplined execution within defined boundaries.

Its SKILL.md is about 1.2k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Development. The repository describes itself as: Memory and Context Orchestration for Coding Agents. The licence is AGPL-3.0.

When your agent uses it

  • Starting a Jumbo goal

Example prompts

  • “/start-jumbo-goal”

Workflow steps

5 steps, taken from the step headings in SKILL.md.

  1. Start the Goal
  2. Understand the Implementation Instructions
  3. Execute Within Scope
  4. Maintain Context as You Implement
  5. Submit for Review

What it can do on your machine

Read from SKILL.md and the folder at commit a02a339. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md (its code samples are bash).

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Start Jumbo Goal loads about 1.2k tokens when it runs. Until then it costs about 46 tokens; SKILL.md has 520 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~46
When it runs · the whole SKILL.md, loaded when a task matches
~1.2k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from jumbocontext/cli at commit a02a339, republished under its AGPL-3.0 licence (© jumbocontext). 520 words, ~1,204 tokens.

Download SKILL.mdSave it as .claude/skills/start-jumbo-goal/SKILL.md (or your agent's skills folder).
name
start-jumbo-goal
description
Use when starting a Jumbo goal. Loads goal context, scope, and architectural constraints, then guides the agent through disciplined execution within defined boundaries.

Start Jumbo Goal

Prompt: Start a Jumbo goal by loading its full context — objective, criteria, scope, architecture, components, decisions, invariants, and guidelines — then execute within the defined boundaries.

Why Structured Implementation Matters

When jumbo goal start is run, it assembles the goal's registered relations into structured implementation instructions. The objective defines what to build. The criteria define how success is measured. The scope defines boundaries. The architectural context (components, decisions, invariants, guidelines) defines constraints. Deviating from these instructions produces work that fails QA review.

Protocol

1. Start the Goal
bash
jumbo goal start --id <goal-id>

This loads the full goal context. Read and internalize every section of the output before writing any code.

2. Understand the Implementation Instructions

The goal start output contains structured sections. Each section carries binding instructions:

Objective

The single purpose of this goal. All work must serve this objective.

Success Criteria

The specific, measurable conditions that determine whether the objective is met. Implementation is not complete until every criterion is satisfied.

Current Progress (if resuming)

If the goal was previously started, review what has already been accomplished and continue from there. Do not redo completed work.

Scope
  • In Scope: Files and areas you MUST work within.
  • Out of Scope: Files and areas you MUST NOT touch.
Solution Architecture
  • Organization Style: Namespaces (directory structures) and file names you introduce MUST maintain this style.
  • Design Patterns: You MUST leverage these patterns where applicable. If the goal requires a pattern not listed, register it with jumbo architecture update.
  • Principles: Artifacts you create MUST directly reflect these principles.
Relevant Components

Components that will be modified, created, or depended upon. Consider these while implementing.

Relevant Dependencies

External libraries involved in the implementation. Consider version constraints and contracts.

Relevant Decisions

Previous architectural decisions that inform or constrain this change. The solution must remain consistent with these decisions.

Invariants

Non-negotiable constraints. You MUST adhere to ALL invariants while implementing.

Guidelines

Coding standards, testing requirements, and process practices. You SHOULD follow these while implementing.

3. Execute Within Scope
  • Work only within the defined scope boundaries.
  • Follow all architectural constraints, invariants, and guidelines.
  • Track progress by documenting completed sub-tasks:
bash
jumbo goal update-progress --id <goal-id> --task-description "<description>"
Show full SKILL.md (235 more words)Show less
4. Maintain Context as You Implement

Register decisions, components, and corrections as they happen — not as a cleanup step after implementation.

  • Architectural decision made (e.g., chose pattern X over Y):

    bash
    jumbo decision add --title "Chose X over Y" --rationale "Because Z" --context "Background"
  • New component created (e.g., introduced a new service or builder):

    bash
    jumbo component add --name "ComponentName" --description "What it does"
    jumbo relation add --from-type goal --from-id <goal-id> --to-type component --to-id <component-id> --type involves --description "Created during implementation"
  • User corrects your approach (e.g., "don't do X, do Y instead"):

    bash
    jumbo invariant add --category "architecture" --description "Never do X because Y"
    jumbo guideline add --category "codingStyle" --description "Prefer X over Y"

Context registered now is served to future sessions. What you skip is lost.

5. Submit for Review

When all success criteria are met and implementation is complete:

bash
jumbo goal submit --id <goal-id>

Rules

  1. Never deviate from the objective. All code changes must serve the stated objective.
  2. Never violate scope boundaries. In-scope files are your workspace. Out-of-scope files are off limits.
  3. Never violate invariants. Invariants are non-negotiable constraints — no exceptions.
  4. Always track progress. Use jumbo goal update-progress to document completed sub-tasks so future agents can resume if needed.
  5. Always submit when done. Do not close or codify the goal yourself. Submit it for QA review.
  6. Do not enter plan mode for refined goals. The goal context IS the plan. Execute immediately.

© jumbocontext, AGPL-3.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in assets/skills/start-jumbo-goal of jumbocontext/cli.

Open the folder on GitHubat commit a02a339

Compare with similar skills

Start Jumbo Goal next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Start Jumbo Goal compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Start Jumbo Goal this skilljumbocontext/cli276—~1.2kAutomated safety check: PassAGPL-3.0
Trellis Session Insightmindfold-ai/Trellis15k4 repos~1.7kAutomated safety check: PassAGPL-3.0
Openspec Verify ChangeFission-AI/OpenSpec71k2 repos~4.6kAutomated safety check: PassMIT
Warp Factory Fileswarpdotdev/warp65k1 repos~2.5kAutomated safety check: PassAGPL-3.0
Migrate Core Code to Submodulestinyhumansai/openhuman42k—~2.6kAutomated safety check: PassGPL-3.0
Analyze Logsactivepieces/activepieces25k1 repos~1.6kAutomated safety check: PassMIT

Similar skills

  • Trellis Session Insight

    mindfold-ai/Trellis

    Reach into past AI conversation history through the trellis mem CLI.

    15k GitHub starsUsed in 4 repos~1.7k tokens
    DevelopmentAuto-check passed
  • Openspec Verify Change

    Fission-AI/OpenSpec

    Verify implementation matches OpenSpec change artifacts. An agent skill from Fission-AI/OpenSpec.

    71k GitHub starsUsed in 2 repos~4.6k tokens
    DevelopmentAuto-check passed
  • Warp Factory Files

    warpdotdev/warp

    Authors and edits file-based Warp software factory definitions rooted at factory.yaml, covering agents, automations, scorers and webhooks, and validates them before a pull request.

    65k GitHub starsUsed in 1 repo~2.5k tokens
    DevelopmentAuto-check passed
  • Migrate Core Code to Submodules

    tinyhumansai/openhuman

    Plans and carries out moving non-host-specific code and its tests from the OpenHuman core into vendored tiny submodule libraries, then releases the submodule and re-pins the host.

    42k GitHub stars~2.6k tokensUpdated yesterday
    DevelopmentAuto-check passed
  • Analyze Logs

    activepieces/activepieces

    Analyze application logs from the .evlog/logs/ directory. An agent skill from activepieces/activepieces.

    25k GitHub starsUsed in 1 repo~1.6k tokens
    DevelopmentAuto-check passed
  • Official

    Runs a loop on a GitHub pull request: fetch review state, triage comments into actions, implement them and resolve threads, repeating until nothing actionable is left.

    48k GitHub stars~2.2k tokensUpdated yesterday
    DevelopmentAuto-check passed

More from jumbocontext/cli

All 13 skills in this repo
  • Codify Jumbo Goal

    jumbocontext/cli

    A skill your agent uses when a Jumbo goal has been approved by QA review and needs architectural reconciliation before closing.

    276 GitHub stars~1.2k tokensUpdated 16 days ago
    Auto-check passed
  • A skill your agent uses when a project has Architecture data that needs migrating to fine-grained entities (Decisions, Invariants, Components, Dependencies).

    276 GitHub stars~881 tokensUpdated 16 days ago
    Auto-check passed
  • Define Jumbo Goals

    jumbocontext/cli

    Add Jumbo goals liberally to decompose objectives into finite units of work with bounded context.

    276 GitHub stars~2.9k tokensUpdated 16 days ago
    Auto-check passed
  • Jumbo Add Guideline

    jumbocontext/cli

    Use liberally when the user expresses a preference about how work should be done — coding style, process, testing approach, communication style.

    276 GitHub stars~522 tokensUpdated 16 days ago
    Auto-check passed
  • Refine Jumbo Goals

    jumbocontext/cli

    A skill your agent uses when Jumbo goals need refinement before implementation.

    276 GitHub stars~1.9k tokensUpdated 16 days ago
    Auto-check passed
  • Reject Jumbo Goal

    jumbocontext/cli

    A skill your agent uses when a Jumbo goal fails QA review and needs to be returned for rework.

    276 GitHub stars~650 tokensUpdated 16 days ago
    Auto-check passed

Questions about Start Jumbo Goal

What does Start Jumbo Goal do?

A skill your agent uses when starting a Jumbo goal. An agent skill from jumbocontext/cli. Start Jumbo Goal is an agent skill from jumbocontext/cli. Use when starting a Jumbo goal.

When should I use Start Jumbo Goal?

Start Jumbo Goal fits situations like: starting a Jumbo goal.

How do I install Start Jumbo Goal in Claude Code?

Run `npx skills add jumbocontext/cli --skill start-jumbo-goal -a claude-code`. Or copy the skill folder (assets/skills/start-jumbo-goal in jumbocontext/cli) into .claude/skills/start-jumbo-goal in your project. Claude Code loads it when a task matches its description.

How do I install Start Jumbo Goal in Codex?

Run `npx skills add jumbocontext/cli --skill start-jumbo-goal -a codex`. Or copy the skill folder (assets/skills/start-jumbo-goal in jumbocontext/cli) into .agents/skills/start-jumbo-goal in your project. Codex loads it when a task matches its description.

Can I use Start Jumbo Goal in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add jumbocontext/cli --skill start-jumbo-goal -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/start-jumbo-goal, .gemini/skills/start-jumbo-goal, .github/skills/start-jumbo-goal and .opencode/skills/start-jumbo-goal in your project.

What does Start Jumbo Goal need to run?

SKILL.md names no scripts, command-line tools or credentials: Start Jumbo Goal is instructions for the agent only.

Does Start Jumbo Goal access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Start Jumbo Goal safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Start Jumbo Goal use?

Start Jumbo Goal is published under the AGPL-3.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Start Jumbo Goal use?

About 1.2k tokens (SKILL.md is roughly 4.8k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Start Jumbo Goal?

Skills that share tags, products or a category with Start Jumbo Goal: Trellis Session Insight (mindfold-ai/Trellis, 15k stars), Openspec Verify Change (Fission-AI/OpenSpec, 71k stars), Warp Factory Files (warpdotdev/warp, 65k stars) and Migrate Core Code to Submodules (tinyhumansai/openhuman, 42k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Start Jumbo Goal?

jumbocontext (a GitHub organization) maintains it in jumbocontext/cli, which has 276 GitHub stars. The repository holds 13 skills in this directory. The repository was last updated on September 23, 2026.

Source: jumbocontext/cli on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.