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
- suiuiui6/harness-engineering-skilltheir own version, first seen Nov 16, 2026
- 0xdewy/skillstheir own version, first seen Nov 16, 2026
- AshrafDurjoy/genai-shared-patternstheir own version, first seen Nov 16, 2026
- hridaya423/devingoaltheir own version, first seen Nov 16, 2026
- OlaPietka/easy-tigertheir own version, first seen Nov 15, 2026
- caidanw/skillstheir own version, first seen Nov 8, 2026
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.
0mlkrizzz655335vSkillorigin
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
PandelisZSkillupdate-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
0mlkrizzz655335vSkillreview-bugbot
缺陷导向代码审查。用 Bugbot 子代理找 bug。
Also in 3 projects that use this
32103210Skillmigrate-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
32103210Skillupdate-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
MicrosoftSkillautomate
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
Other versions50 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 50
- PrimeIntellect-ai/prime-agent★ 23ksame name
- coco-research/coco★ 797same name
- cwinvestments/memstack★ 433same name
- jthack/claude-goal★ 113same name
- svy04/ballast★ 73same name
- anthony-chaudhary/fak★ 49same name
- lennoxsaint/Skills★ 48same name
- Infinite-Labs-AI/infinite-skills★ 43same name
- manusco/resonance★ 38same name
- Sanexxxx777/curated-claude-code★ 16same name
40 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository14 other files it gives its agents.
- .claude/AGENTS.mdAGENTS.md
- .claude/CLAUDE.mdCLAUDE.md
- alphaxivSkill
- askSkill
- autoexploreSkill
- collateSkill
- curriculumSkill
- editSkill
- notionSkill
- obsidianSkill
- retrieveSkill
- tutorSkill
- webSkill
- zoteroSkill
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
- Very specific: line 22
If $ARGUMENTS starts with edit , extract the slug and jump to Edit Mode.
- Distinctive: line 23
If $ARGUMENTS is empty, ask the user for a topic, then continue with Step 1.
- Very specific: line 32
Read course/<slug>/goal.md and course/<slug>/progress.md (if present).
- Distinctive: line 60
Ask for anything important that's missing before continuing.
- Very specific: line 87
If course/<slug>/progress.md exists (written by /tutor), read its Item Progress table and map each scored item to its retrieved file, so …
- Very specific: line 115
Create course/<slug>/ if needed and write course/<slug>/goal.md from .claude/Templates/goal.md.
- Very specific: line 138
If they decline, point them to run /curriculum <slug> themselves when ready, and to /goal edit <slug> to adjust the goal.
“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.
Files
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 asYYYY-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;
/curriculumschedules 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:
| Status | Rung — what it means | Condition |
missing | 0 — 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 |
pulled | 1 — 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 |
untested | 2 — 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) |
revisit | 2 — 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 |
covered | 3 — 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:
- Tell the user you'll now build the curriculum, and that
/curriculumwill 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. - Unless they decline, invoke
/curriculum <slug>straight away (it reads the goal you just wrote and continues the pipeline through to an initializedprogress.md). - If they decline, point them to run
/curriculum <slug>themselves when ready, and to/goal edit <slug>to adjust the goal.
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
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.

