agentleFS
sebastientaggartsebastientaggart/CodeCannon/skills/github-agile/delegate/SKILL.md

delegate

Skill1 starsChanged 8 months agoMITView on GitHub

Code Cannon: Generate a scoped prompt for delegating work to another agent

78 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use state, preview and 9 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • MicrosoftSkill

    state

    Store and query a Microsoft Managed App's built-in state and file attachments with `ms project state` CLI schema commands and generated TypeScript clients. First, with the user signed in to the CLI, run `ms feature status --name state --json`; if `feature.enabled` is not `true`, recommend an alternative connector or data source.

    Also in 3 projects that use this

  • kv0906Skill

    preview

    Render vault notes as styled HTML and open in browser. Turn markdown docs, decisions, progress reports into readable previews. Use for "/preview docs/project-a/checkout-flow.md", "/preview progress all", "/preview decide auth-approach", or when user wants to see a doc before sharing.

    Also in 3 projects that use this

  • GitHubSkill

    prd

    Generate high-quality Product Requirements Documents (PRDs) for software systems and AI-powered features. Includes executive summaries, user stories, technical specifications, and risk analysis.

    Also in 5 projects that use this

  • nicolasestremSkill

    document

    Create focused documentation for specific components or features

    Also in 3 projects that use this

  • anombyte93Skill

    discover

    Phase 1 of the prd-taskmaster pipeline: brainstorm-driven discovery. Delegates to superpowers:brainstorming in Interactive Mode (one adaptive question at a time), or self-brainstorms in Autonomous Mode when no user is present. Intercepts before the brainstorming chain hands off to writing-plans — this skill owns the exit. Extracts constraints, calibrates scale (Solo / Team / Enterprise), and advances the pipeline to GENERATE.

    Also in 3 projects that use this

  • Microsoft AzureSkill

    skill-authoring

    Write Agent Skills that comply with the agentskills.io specification. WHEN: "create a skill", "new skill", "write a skill", "skill template", "skill structure", "review skill", "skill PR", "skill compliance", "SKILL.md format", "skill frontmatter", "skill best practices". DO NOT USE FOR: general documentation. INVOKES: vally CLI.

    Also in 3 projects that use this

The file

We split this skill into the rules it gives an agent

7 rules. 0 are common, 3 are distinctive, and 4 are very specific to this project.

  • 0

    Common

    Most of the 67 edited versions we read keep these.

  • 3

    Distinctive

    This author's own choices. Worth a look.

  • 4

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    If $ARGUMENTS is empty, ask: "What task should the prompt describe?" and wait for a response.

    line 24
  2. Very specific:

    If that errors because the {{BRANCH_DEV}} ref isn't present locally, diff against {{BRANCH_PROD}}...HEAD instead.

    line 43
  3. Very specific:

    Use the issue title, body, and agent implementation notes from the current /start session — do not re-fetch them.

    line 52
  4. Very specific:

    After presenting, record internally that $ARGUMENTS has been delegated.

    line 94
  5. Distinctive:

    Never commit, push, or modify any files.

    line 104
  6. Distinctive:

    Never post to GitHub (no comments, labels, or issue updates).

    line 105
  7. Distinctive:

    Do not suggest or name a specific agent to paste into — that is the operator's choice.

    line 107

“Common” is measured: kept by at least 6 in 10 of the 67 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

What /delegate does

/delegate generates a ready-to-paste prompt that gives another AI agent everything it needs to contribute code on the current branch. Run it during the planning gate of /start (Step 2), before saying go, to carve off part of the work for another agent.

Each /delegate invocation also tells the lead agent to skip that work when it proceeds.


Prerequisites

/delegate relies on context already loaded by /start. If you are not in the middle of a /start planning gate (Case A Step 2 or Case B Step 2), say:

"/delegate is meant to be run during the planning step of /start, before you type go. Run /start first to load the issue context, then use /delegate to carve off tasks."

Stop.

If $ARGUMENTS is empty, ask: "What task should the prompt describe?" and wait for a response.


Step 1 — Gather branch and diff context

git branch --show-current
{{#if BRANCH_DEV}}
git diff --name-only {{BRANCH_DEV}}...HEAD
{{/if}}
{{#if !BRANCH_DEV}}
git diff --name-only {{BRANCH_PROD}}...HEAD
{{/if}}

{{#if BRANCH_DEV}} If that errors because the {{BRANCH_DEV}} ref isn't present locally, diff against {{BRANCH_PROD}}...HEAD instead. {{/if}}


Step 2 — Build the prompt

Compose the following as a fenced markdown block (```markdown) so the operator can copy it.

Use the issue title, body, and agent implementation notes from the current /start session — do not re-fetch them.

## Your task

<$ARGUMENTS — the operator's description of the delegated work>

## Branch

You are on branch `<branch name>`. Stay on this branch.

## Context

**Issue #<number>: <title>**

<issue body — include the full body as-is>

<if there are agent implementation notes from the issue comments, include them under a "### Agent notes" subheading>

## What's already been done on this branch

<list of changed files from git diff, or "No changes yet — clean branch." if empty>

## Rules

- **Do not commit.** Leave all changes unstaged. The lead developer will handle commits.
- **Be surgical.** Only modify what is necessary for your task. Do not refactor adjacent code, reorganize imports, or "improve" files outside the scope of your task.
- **Do not start until the branch exists on the remote.** The lead developer will create and push the branch. Wait until `git checkout <branch name>` followed by `git pull origin <branch name>` succeeds before beginning work.

Step 3 — Present to operator and update lead scope

Print the generated prompt preceded by:

Here's a delegation prompt for: $ARGUMENTS

Review and edit as needed, then paste into your agent of choice after the lead agent has created and pushed the branch.

Then print the fenced prompt block.

After presenting, record internally that $ARGUMENTS has been delegated. When the operator says go and the lead agent proceeds to write code, it must skip the delegated work. If multiple /delegate invocations occur before go, accumulate all of them.

After presenting, say:

Noted — the lead agent will skip this part when you say go. Run /delegate <another task> to delegate more, or type go to proceed.


Hard rules

  • Never commit, push, or modify any files.
  • Never post to GitHub (no comments, labels, or issue updates).
  • Output only — the skill produces text for the operator to copy.
  • Do not suggest or name a specific agent to paste into — that is the operator's choice.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.