agentleFS
Agents4Academia-AIAgents4Academia-AI/UReKA/.claude/skills/goal/SKILL.md

goal

Skill9 starsChanged 5 months agoMITView on GitHub

Set a learning goal — define the target and scope, audit your knowledge base against it, and classify what you already know vs what must be pulled in. First stage of a course; hands off to /curriculum.

50 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use origin, update-cli-config 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.
  • 0mlkrizzz655335vSkill

    origin

    origin CLI 管理。安装、登录、更新、修复 origin.cursor.com 的 CLI。"disabled-environments:

    Also in 3 projects that use this

  • PandelisZSkill

    update-cli-config

    View and modify Cursor CLI configuration settings in ~/.cursor/cli-config.json. Use when the user wants to change CLI settings, configure permissions, switch approval mode, enable vim mode, toggle display options, configure sandbox, or manage any CLI preferences.

    Also in 4 projects that use this

  • 0mlkrizzz655335vSkill

    review-bugbot

    缺陷导向代码审查。用 Bugbot 子代理找 bug。

    Also in 3 projects that use this

  • 32103210Skill

    migrate-to-skills

    Convert 'Applied intelligently' Cursor rules (.cursor/rules/*.mdc) and slash commands (.cursor/commands/*.md) to Agent Skills format (.cursor/skills/). Use when you want to migrate rules or commands to skills, convert .mdc rules to SKILL.md format, or consolidate commands into the skills directory.

    Also in 4 projects that use this

  • 32103210Skill

    update-cursor-settings

    Modify Cursor/VSCode user settings in settings.json. Use when you want to change editor settings, preferences, configuration, themes, font size, tab size, format on save, auto save, keybindings, or any settings.json values.

    Also in 4 projects that use this

  • MicrosoftSkill

    automate

    Propose an automation from a completed task or a new idea, then create it after approval. Use when the user asks to create an automation or wants scheduled local or cloud agent work.

    Also in 4 projects that use this

  • +3 more from 0mlkrizzz655335v
  • +1 more from 32103210

The file

We split this skill into the rules it gives an agent

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

  • 0

    Common

    Most of the 46 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 5

    Very specific

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

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

    If $ARGUMENTS starts with edit , extract the slug and jump to Edit Mode.

    line 22
  2. Distinctive:

    If $ARGUMENTS is empty, ask the user for a topic, then continue with Step 1.

    line 23
  3. Very specific:

    Read course/<slug>/goal.md and course/<slug>/progress.md (if present).

    line 32
  4. Distinctive:

    Ask for anything important that's missing before continuing.

    line 60
  5. Very specific:

    If course/<slug>/progress.md exists (written by /tutor), read its Item Progress table and map each scored item to its retrieved file, so …

    line 87
  6. Very specific:

    Create course/<slug>/ if needed and write course/<slug>/goal.md from .claude/Templates/goal.md.

    line 115
  7. Very specific:

    If they decline, point them to run /curriculum <slug> themselves when ready, and to /goal edit <slug> to adjust the goal.

    line 138

“Common” is measured: kept by at least 6 in 10 of the 46 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

Set the goal for a course on $ARGUMENTS: define the target and scope, audit the knowledge base against it, classify what's already known vs missing, and write course/<slug>/goal.md. This is the first stage of the learning pipeline:

/goal <topic>   →  course/<slug>/goal.md   (target + scope + knowledge audit)
                   └─ auto-hands off to ↓ (goal + curriculum are linked)
/curriculum <slug>  →  pulls resources, builds modules + schedule
/tutor <slug>       →  daily: reading list → teach → ingest → (test|honor) → advance

/goal does not pull resources or build a schedule — that is /curriculum's job; keep this stage about what to learn and where you stand. The two stages are linked: once the goal is confirmed, /goal auto-hands off to /curriculum <slug> (see Wrap up).

If $ARGUMENTS starts with edit , extract the slug and jump to Edit Mode. If $ARGUMENTS is empty, ask the user for a topic, then continue with Step 1.


Edit Mode — /goal edit <slug>

Adjust an existing goal without rebuilding from scratch.

E1 — Load

Read course/<slug>/goal.md and course/<slug>/progress.md (if present). Show the current state: target, scope, subgoals, knowledge audit.

E2 — Ask what to change

Offer (user may pick several): change deadline / hours; tighten or widen scope; add/edit/remove a subgoal; re-run the audit for a new sub-topic; add free-form notes.

E3 — Apply changes

Make the edits. If a new sub-topic is added, re-run retrieval (Step 3) for just that topic and fold the findings into the audit. Do not touch plan.md/schedule.md — if the scope changed materially, tell the user to re-run /curriculum <slug>.

E4 — Write

Overwrite course/<slug>/goal.md. Report what changed.


Step 1 — Gather inputs

Parse $ARGUMENTS to extract:

  • topic — what to learn (required; ask if missing)
  • deadline — target date, e.g. "by 15 July", "in 4 weeks" (optional; convert relative phrases to absolute dates with date +%Y-%m-%d; store as YYYY-MM-DD; use "open-ended" if not given)
  • hours_per_week — available study hours per week (optional; suggest 5 default). This is the only time budget — do not assume a per-day rate or split the week into days; /curriculum schedules in hour-blocks within each week.

Ask for anything important that's missing before continuing.

Step 2 — Slug and existing check

Slugify the topic to kebab-case. Check whether course/<slug>/goal.md exists.

  • Exists → ask: update it (→ Edit Mode) or start fresh (overwrite)?
  • Does not exist → continue.

Step 3 — Scope the topic

Draft an explicit in-scope / out-of-scope definition (e.g. for flow matching: in — CNFs, the FM objective, probability paths, sampling; out — GANs/VAEs except as contrast, deployment). Show it to the user and refine before auditing — the scope is what /curriculum will pull resources against, so it must be right.

Step 4 — Audit the knowledge base

sh .claude/src/pyrun .claude/src/retrieve_cli.py --retrieve "<topic>"

Each output line is <file path>\t<title>. Read each result and follow relevant [[wikilinks]] (1–2 hops) with sh .claude/src/pyrun .claude/src/retrieve_cli.py --resolve "<link>". Collect relevant files across sources/, notes/, papers/, concepts/.

Step 5 — Read existing scores

If course/<slug>/progress.md exists (written by /tutor), read its Item Progress table and map each scored item to its retrieved file, so the audit reflects tested mastery, not just presence.

Step 6 — Classify knowledge state

Status is a single readiness ladder — how ready each item is in your base. Assign each item the highest rung it has reached:

StatusRung — what it meansCondition
missing0 — not in your base. In scope but you have nothing on it yet; /curriculum must pull a source.In scope / referenced (wikilinks, concepts_mentioned) but no source or page exists
pulled1 — raw source only. A source sits in sources//notes/, but it hasn't been written up into a page yet.Source exists in sources//notes/, no synthesised papers//concepts/ page
untested2 — page exists, untested. You have a synthesised page but /tutor hasn't tested you on it.Page exists in papers//concepts/, no score yet (incl. honor-system read)
revisit2 — page exists, tested but shaky. Tested, but you scored partial/wrong — go back to it.Page exists and latest /tutor score is partial or wrong
covered3 — page exists, tested solid. The top rung: a page exists and you tested confident.Page exists and latest score is confident

The ladder encodes two axes: (a) is it written up? (missing/pulled = no page; untested/revisit/covered = a page exists) and (b) has /tutor tested it? (untested = not yet; revisit/covered = tested).

Step 7 — Target and subgoals

Draft ## Target (one sentence: what the learner will understand or be able to do by the deadline) and ## Subgoals (one per thematic cluster of the scope).

Step 8 — Write goal.md

Create course/<slug>/ if needed and write course/<slug>/goal.md from .claude/Templates/goal.md. Frontmatter type: goal, hours_per_week, tutor_prompt_mode: unset, testing_mode: unset (set later by /tutor to test | honor), coverage_pct: 0 (no hours_per_day — the week is the only budget). Knowledge Audit is the 6-column table (Item | Type | Status | Score | Last tested | Source); untested items carry — for Score/Last tested.

coverage_pct and the Score/Last tested columns are kept in sync by /tutor; leave them at their initial values here.

Wrap up

Report: course/<slug>/goal.md written; counts of covered / revisit / untested / pulled / missing items.

Then auto-hand off to /curriculum — a goal and its curriculum are linked, so do not stop at the goal:

  1. Tell the user you'll now build the curriculum, and that /curriculum will pull vetted resources (papers + Wikipedia + blogs) into this course's library — the most expensive step. Give them a chance to adjust the scope first (/goal edit <slug>) or to skip the auto-handoff if they only wanted the goal for now.
  2. Unless they decline, invoke /curriculum <slug> straight away (it reads the goal you just wrote and continues the pipeline through to an initialized progress.md).
  3. If they decline, point them to run /curriculum <slug> themselves when ready, and to /goal edit <slug> to adjust the goal.

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.