agentleFS
nesnilnehcnesnilnehc/ai-cortex/skills/review-tasks/SKILL.md

review-tasks

Skill8 starsChanged 2 months agoMITView on GitHub

Review an existing task list against its modeling Spec and canonical quality Rule before assignment or coding, including design coverage, affected scope, Rule references and verification evidence.

2 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈7

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    After task derivation and before assignment or coding

    line 26
  2. Distinctive:

    After design changes alter task coverage or dependencies

    line 27
  3. Distinctive:

    Before an agent starts a quality-sensitive implementation task

    line 28
  4. Distinctive:

    Do not invent tasks, dependencies, owners, estimates or verification commands.

    line 46
  5. Distinctive:

    Do not mark tasks complete.

    line 47
  6. Distinctive:

    Do not copy engineering Rule text into a task.

    line 48
  7. Distinctive:

    If the technical design has no engineering quality mapping and the task changes one document, do not require governance annotations.

    line 67

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “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

Skill: Review Tasks

Purpose

Evaluate an existing task list against task-modeling and task-quality. This atomic pre-coding gate ensures implementation work preserves design traceability and verification obligations.

Core objective

Determine whether every task is independently executable, dependency-safe, bounded, traceable to design and explicit about applicable engineering checks.

Scope boundaries

This Skill reviews one task-list artifact and its technical-design parent. It does not schedule work, assign people, implement tasks, run tests or aggregate other reviews.

Use cases

  • After task derivation and before assignment or coding
  • After design changes alter task coverage or dependencies
  • Before an agent starts a quality-sensitive implementation task

Behavior

  1. Load task-quality in full, record its version, and resolve the approved technical-design parent.
  2. Resolve applicability per rule-modeling: whether the list is being handed off, which tasks meet an engineering-governance trigger, any declared parameters, and any valid waivers.
  3. Evaluate each applicable item independently, gathering only the evidence that item declares. A missing declared parameter makes its item evidence-limited; it never suppresses a baseline item.
  4. Emit one finding per failed obligation, with category task-quality, a task ID or document-level location, and the failed Rule ID in the description, for example task-quality@<active-version>/TASK-004.
  5. Apply a waiver only when every waiver field is valid and its Rule ID and scope cover the exact finding. Report waived items separately; do not erase their existence.
  6. Return Rule coverage using the exact passed, waived, not_applicable and evidence_limited fields from the findings-list Spec.
  7. Report the list ready for assignment only after every applicable item has a pass, a valid waiver or an explicit evidence limitation; do not start implementation.

Input and output

Input is one task-list artifact plus its local technical-design parent. Output follows findings-list: every finding uses category task-quality and cites the failed Rule ID, and the coverage footer is metadata rather than a finding.

Restrictions

  • Do not invent tasks, dependencies, owners, estimates or verification commands.
  • Do not mark tasks complete.
  • Do not copy engineering Rule text into a task; require stable Rule references.

Self-Check

  • The active Spec and Rule were loaded.
  • Parent status, DAG and design coverage were checked.
  • Conditional engineering governance annotations were checked.
  • Every finding names a precise task or section, uses category task-quality and cites one failed Rule ID.
  • Every applicable Rule ID received a pass, a finding, a valid waiver or an explicit evidence limitation.
  • The coverage footer distinguishes passed, waived, not applicable and evidence limited.

Examples

Example 1: cross-module task has no checks

A task changes two declared modules but has no affected-scope or ARC verification annotation. Emit a task-quality finding; do not decide the missing topology.

Example 2: local documentation task

If the technical design has no engineering quality mapping and the task changes one document, do not require governance annotations.

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.