agentleFS
adezdevadezdev/blueprint/skills/tasks/SKILL.md

tasks

Skill0 starsChanged 3 months agoMITView on GitHub

Turn an architecture.md into a traceable tasks.md — phased, commit-sized T-ids, each citing its component and acceptance criteria with a runnable done-when check. Use when the user says "/blueprint:tasks", "break this into tasks", "make a task list", "what's the build order", or has an architecture.md and wants an executable plan.

49 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use send-user-message, tinyagi-admin and 10 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.
  • Maliot100XSkill

    send-user-message

    Send a proactive message to a paired user via their channel (Discord, Telegram, or WhatsApp). Use when the agent needs to notify, alert, or send an unsolicited message to a user — especially during heartbeat invocations, scheduled tasks, or when the agent wants to reach out without a prior user message in the current conversation. Triggers: 'send message to user', 'notify user', 'alert user', 'message the user on discord/telegram/whatsapp', or any need to proactively communicate with a paired sender.

    Also in 4 projects that use this

  • jlia0Skill

    tinyagi-admin

    Manage and operate the TinyAGI system itself — agents, teams, settings, queue, tasks, daemon lifecycle, and source code. Use when the agent needs to: list/add/remove/update agents or teams, check queue status, view logs, start/stop/restart TinyAGI, change settings (provider, model, channels), send messages to the queue, manage tasks, retry dead-letter messages, view recent responses, modify TinyAGI source code or configuration, or perform any administrative operation on the TinyAGI platform. Triggers: 'manage tinyagi', 'add an agent', 'remove a team', 'check queue', 'view logs', 'restart tinyagi', 'change provider', 'update settings', 'create a task', 'modify tinyagi code'.

    Also in 3 projects that use this

  • ArvorCoSkill

    constitution

    Create or update the project constitution - personalized governance and coding principles. Use when starting a project or updating project standards. Triggers on: create constitution, update principles, set project rules.

    Also in 4 projects that use this

  • ArvorCoSkill

    specify

    Create feature specification from natural language description. Use when starting a new feature or creating a spec. Triggers on: create spec, specify feature, new feature spec.

    Also in 6 projects that use this

  • GoogleSkill

    quality

    Evaluates whether a GitHub issue is spam, empty, needs more information, or is OK to proceed.

    Also in 3 projects that use this

  • anyprotoSkill

    skills-manager

    Context-aware routing to skills and hooks management. Use when troubleshooting skill activation, fine-tuning keywords, or managing the automated documentation system.

    Also in 3 projects that use this

  • +1 more from ArvorCo

The file

We split this skill into the rules it gives an agent

7 rules. 0 are common, 5 are distinctive, and 2 are very specific to this project.

  • 0

    Common

    Most of the 42 edited versions we read keep these.

  • 5

    Distinctive

    This author's own choices. Worth a look.

  • 2

    Very specific

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

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

    Read ${CLAUDE_PLUGIN_ROOT}/FORMAT.md first (or FORMAT.md at the repo root if you are working inside the blueprint repo itself).

    line 8
  2. Distinctive:

    Never emit a "write all the tests" phase.

    line 36
  3. Distinctive:

    Look for dependencies between *shapes*, not only between tasks.

    line 46
  4. Very specific:

    Cover every AC: Before writing the file, check both directions: every R<n>.AC<m> in requirements.md is cited by at least one task, and …

    line 55
  5. Distinctive:

    No task larger than a reviewable commit.

    line 77
  6. Distinctive:

    No dedicated "testing" or "documentation" phase at the end.

    line 79
  7. Distinctive:

    No task that exists only because it seemed like good hygiene.

    line 80

“Common” is measured: kept by at least 6 in 10 of the 42 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

Blueprint: tasks

Read ${CLAUDE_PLUGIN_ROOT}/FORMAT.md first (or FORMAT.md at the repo root if you are working inside the blueprint repo itself). It defines the ID grammar and the exact document shape. Do not improvise a variant.

Input

.blueprint/<slug>/architecture.md and requirements.md — you need both, because tasks cite acceptance criteria, not just components. Resolve the slug from the argument, the single existing feature directory, or ask.

What you do

  1. One task per commit-sized change. Roughly ≤3 files. A task a reviewer cannot hold in their head in one sitting is two tasks.
  2. Cite the chain on every task: → C2 | R1.AC1, R1.AC2. This is what makes /blueprint:build able to make a focused change — it reads only the cited component and criteria, not the whole spec. A task with no citation is either a chore (→ chore) or a task nobody asked for.
  3. Every task gets a done-when that is a command. pytest tests/x.py -k y passes, curl localhost:8000/health returns 200, npm run build succeeds. If a task genuinely has no automatable check, write the exact manual observation — but treat that as a smell, and prefer restructuring so it has one.
  4. Tests are part of the task, not a phase at the end. The task that implements R1.AC1 is the task that writes the test asserting R1.AC1. Never emit a "write all the tests" phase.
  5. Phase by shippability, not by layer. Each phase should end somewhere the branch could be reviewed and merged. "All models, then all handlers, then all routes" is a layer plan and produces one enormous unreviewable PR. Prefer thin vertical slices.
  6. Order by dependency, mark it explicitly. (after T1). Tasks with no after line are unblocked and may be done in any order.

    Look for dependencies between shapes, not only between tasks. A task often needs a type, constant or helper that its own component does not declare — especially where components share a file. Read each task's component interface, note what it uses but does not define, and find the component that owns it. Then either order that component's task first, or say in the task that it introduces the shape and which component owns it. Leaving it implicit means whoever runs /blueprint:build hits it mid-task and has to stop.

  7. Cover every AC. Before writing the file, check both directions: every R<n>.AC<m> in requirements.md is cited by at least one task, and every C<n> in architecture.md is implemented by at least one task. Report any that are not — do not paper over the gap.
  8. Write the file to .blueprint/<slug>/tasks.md.

Updating an existing tasks.md

Append. Never renumber, never rewrite a completed [x] task. If a task is no longer needed, leave it unchecked and add status: dropped — <reason>.

Done

Report: task count, phase count, coverage check (any uncovered AC or unimplemented component?), and the first unblocked task. Then:

Next: /blueprint:inspect <slug> to validate the chain, then /blueprint:build <slug> to execute T1.

Rules

  • No task larger than a reviewable commit.
  • No done-when: code is written. That is not a check.
  • No dedicated "testing" or "documentation" phase at the end.
  • No task that exists only because it seemed like good hygiene. If no component or criterion asks for it, cut it.

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.