task-observer
Skill3.9k starsChanged 53 days agoCC-BY-4.0View on GitHub
Turns corrections, repeated workflows and judgement calls from work sessions into small, evidenced skill updates: logs each observation with its source, stages every change for your approval, and flags repeating work as new skills. Use during ANY multi-step task, agentic workflow, or work session. Also triggers in post-task feedback discussions and when the user mentions skill observations, the observation log, or skill taxonomy. Also known as \"One Skill to Rule Them All\" — trigger on this phrase too. IMPORTANT: invoke this skill before the FIRST tool call of any session and before writing or proposing a plan — any turn that will involve a tool call counts. This sentence is the session-start trigger and the only activation layer that survives an unreachable config file; pair it with a CLAUDE.md instruction or a session-start hook (references/environments.md). A subagent dispatched by a session already running it does not run it: it writes nothing and puts its findings in its report.
11 other projects use a version of this, 5 of them edited; 5 of 5 edited versions we compared changed its rules; 1 new project in the last 4 weeks, against 0 the 4 before.
How people change it
5 of 5 edited versions changed its rules, and some changes recur.
Of the 5 edited versions we compared: 5 made changes of their own.
Rules people drop
- A rule we can't quote: its licence doesn't allow itdropped by 5 of 5
- A rule we can't quote: its licence doesn't allow itdropped by 5 of 5
- A rule we can't quote: its licence doesn't allow itdropped by 4 of 5
- A rule we can't quote: its licence doesn't allow itdropped by 3 of 5
- A rule we can't quote: its licence doesn't allow itdropped by 3 of 5
Rules people add
- Scan. Read only the frontmatter of each file in
observation-log/— the header block between the first two---lines, never the bodies — and build awareness fromstatus,skill,proposes_skillandtitle; also read the active principles. …added by 5 of 5 - Review trigger. Read
skill-observations/last-review-date.txt. The value carries the truth: a date = when the last review actually ran;never= no review has run yet. …added by 5 of 5 - Context preservation: if an observation depends on session-local data (uploads, API output), save that context into the workspace first and set
reference:to its path — an observation whose evidence dies with the session is incomplete.added by 5 of 5 - Create
skill-observations/last-review-date.txtcontaining the literal valueneverif it doesn't exist — never write a date into it at setup; a date means a review actually ran. …added by 4 of 5 - First run. If the log is empty and the project has history (handover or decision docs, commit history, test scripts, an existing CLAUDE.md — which is largely a record of corrections nobody logged), offer a one-off backfill pass over those artefacts. …added by 4 of 5
- Storage. If
skill-observations/observation-log/(with itsarchive/subdirectory) orskill-observations/cross-cutting-principles.mddon't exist, create them (principles template:references/skill-authoring.md). …added by 3 of 5 - If
skill-updates/PENDING.mdlists staged updates, say "N staged updates awaiting review" in one line.added by 3 of 5 - Batch writes: resolve each id at its own write time. When logging more than one observation in a session that may overlap a scheduled review or another writer, run the id snippet before EACH file — never pre-compute a range and hardcode sequential numbers …added by 3 of 5
We split this file and each edited version into the rules they give an agent and match them by wording. A version that adds no rule of its own and has not changed since this file last did reads as an older text, so what it lacks is not counted as dropped. We keep no past texts of a file, so that is inferred, and an edited version that started from an older text still counts its gaps as drops.
Adopted by
Surging: 1 new project took it in the last 4 weeks, none the 4 weeks before.thin data
Most recently taken up by
- willy92wins/DayZ-Modding-Knowledge-Packedited, first seen Nov 8, 2026
- VorElze/hermes-office-skillsedited, first seen Nov 1, 2026
- mkarami313/Claude-skilltheir own version, first seen Nov 1, 2026
- okfriansyah-moh/agentic-workflowtheir own version, first seen Oct 25, 2026
- jeremybrasher/grokbot-skillsedited, first seen Oct 25, 2026
- beelabstudio/aiedited, first seen Sep 20, 2026
Who's talking about rebelytics/one-skill-to-rule-them-all
- A Hacker News story“Show HN: A Claude meta-skill that improves all your skills, including itself”Hacker News · Feb 16, 2026 · 1 points
- sudosubin/agents.nix“claude-code-marketplaces.github.rebelytics.one-skill-to-rule-them-all: 3.10.0 → 3.11.0”GitHub · Nov 15, 2026
- sudosubin/agents.nix“claude-code-plugins.github.rebelytics.one-skill-to-rule-them-all: 3.10.0 → 3.11.0”GitHub · Nov 15, 2026
- sudosubin/agents.nix“agent-skills.github.rebelytics.one-skill-to-rule-them-all: 3.10.0 → 3.11.0”GitHub · Nov 15, 2026
- Shogo-nfrealmusic/Knot-AI“free/67: Claude Code の 5 つのプラグインのページ(67-claude-code-plugins)”GitHub · Nov 14, 2026
The biggest 5 of 43 mentions linking to the repository that our crawl found. Each line is their own headline.
See all 43
Talked about on GitHub and Hacker News (43 times).
On GitHub · 42
- Nov 15, 2026claude-code-marketplaces.github.rebelytics.one-skill-to-rule-them-all: 3.10.0 → 3.11.0 — sudosubin/agents.nix0 reactions
- Nov 15, 2026claude-code-plugins.github.rebelytics.one-skill-to-rule-them-all: 3.10.0 → 3.11.0 — sudosubin/agents.nix0 reactions
- Nov 15, 2026agent-skills.github.rebelytics.one-skill-to-rule-them-all: 3.10.0 → 3.11.0 — sudosubin/agents.nix0 reactions
- Nov 14, 2026free/67: Claude Code の 5 つのプラグインのページ(67-claude-code-plugins) — Shogo-nfrealmusic/Knot-AI0 reactions
- Nov 13, 2026skills/skill-inspector: step 1 reviews the default-branch clone, not the artefact that gets installed — NVIDIA/SkillSpector0 reactions · 1 comment
37 more on GitHub
- Nov 9, 2026Enter animations that persist their end state (fill-mode both/forwards) leave an identity transform that traps position: fixed descendants — emilkowalski/skills0 reactions
- Nov 9, 2026Page transitions: a persisted end keyframe (fill-mode both/forwards) leaves an identity transform that traps position: fixed children — LottieFiles/motion-design-skill0 reactions
- Nov 9, 2026[skill] hyperframes-cli: unattended `npx hyperframes` stops at npm's "Ok to proceed?" prompt, and a pinned call can still fail in npx's cache step — heygen-com/hyperframes0 reactions · 2 comments
- Nov 7, 2026claude-code-marketplaces.github.rebelytics.one-skill-to-rule-them-all: 3.9.0 → 3.10.0 — sudosubin/agents.nix0 reactions
- Nov 6, 2026claude-code-plugins.github.rebelytics.one-skill-to-rule-them-all: 3.9.0 → 3.10.0 — sudosubin/agents.nix0 reactions
- Nov 6, 2026agent-skills.github.rebelytics.one-skill-to-rule-them-all: 3.9.0 → 3.10.0 — sudosubin/agents.nix0 reactions
- Nov 5, 2026skills: port task-observer (CC-BY-4.0, modified) and systematic-debugging (MIT) — willy92wins/DayZ-Modding-Knowledge-Pack0 reactions
- Nov 2, 2026agent-skills.github.rebelytics.one-skill-to-rule-them-all: 3.8.0 → 3.9.0 — sudosubin/agents.nix0 reactions
- Nov 2, 2026claude-code-marketplaces.github.rebelytics.one-skill-to-rule-them-all: 3.8.0 → 3.9.0 — sudosubin/agents.nix0 reactions
- Nov 2, 2026claude-code-plugins.github.rebelytics.one-skill-to-rule-them-all: 3.8.0 → 3.9.0 — sudosubin/agents.nix0 reactions
- Oct 28, 2026claude-code-plugins.github.rebelytics.one-skill-to-rule-them-all: init at 3.8.0 — sudosubin/agents.nix0 reactions
- Oct 28, 2026claude-code-marketplaces.github.rebelytics.one-skill-to-rule-them-all: init at 3.8.0 — sudosubin/agents.nix0 reactions
- Oct 23, 2026agent-skills.github.rebelytics.one-skill-to-rule-them-all: 3.7.0 → 3.8.0 — sudosubin/agents.nix0 reactions
- Oct 21, 2026claude.ai skill sync stopped completing on 2026-09-30: an updated skill shows its new description but keeps the old files on disk (Windows) — anthropics/claude-code0 reactions
- Oct 19, 2026[Bug] MCP obsidian_vault_health reports 10 wanted notes on a fresh bootstrap: Home.md links use \| outside a table, and folder links pass unseen — eugeniughelbur/obsidian-second-brain0 reactions
- Oct 16, 2026agent-skills.github.rebelytics.one-skill-to-rule-them-all: 3.6.0 → 3.7.0 — sudosubin/agents.nix0 reactions
- Oct 14, 2026agent-skills.github.rebelytics.one-skill-to-rule-them-all: 3.5.0 → 3.6.0 — sudosubin/agents.nix0 reactions
- Oct 12, 2026[Candidate] rebelytics/one-skill-to-rule-them-all - task-observer — ProSkillsMD/proskills0 reactions · 1 comment
- Oct 10, 2026Add task-observer skill via skills CLI — lynkzyck/SQL-Syntax0 reactions
- Oct 9, 2026Add project skills (installed only): task-observer, impeccable, four from emilkowalski/skills — civilengineer117612-png/fo-und0 reactions
- Oct 4, 2026Update skill: /rebelytics-task-observer (critical risk) — aiskillstore/marketplace0 reactions
- Oct 4, 2026agent-skills.github.rebelytics.one-skill-to-rule-them-all: 3.3.1 → 3.5.0 — sudosubin/agents.nix0 reactions
- Oct 1, 2026agent-skills.github.rebelytics.one-skill-to-rule-them-all: init at 3.3.1 — sudosubin/agents.nix0 reactions
- Sep 29, 2026Admin AI: Install skills https://github.com/mattpocock/skills https://github.com/ — mehrshadevilb-cell/ArtistYar-Website0 reactions
- Sep 29, 2026Admin AI: Install skills https://github.com/mattpocock/skills https://github.com/ — mehrshadevilb-cell/ArtistYar-Website0 reactions
- Sep 29, 2026관찰자(observer) 에이전트 / skill 개선 outer loop (+ task-observer) — 적대적 검토·기록 — chonamdoo/agent-flow0 reactions
- Sep 27, 2026Strengthen wrap-up and session lesson learning loop — mhingston/agent-skills0 reactions
- Sep 18, 2026Add Task Observer to Development & Code Tools — ComposioHQ/awesome-claude-skills0 reactions
- Sep 18, 2026Add task-observer (self-improving meta-skill) to Development Tools — JackyST0/awesome-agent-skills0 reactions · 1 comment
- Sep 6, 2026[CE] Native Headroom (max token savings) + Task Observer–informed self-improvement skill — ShaftHQ/SHAFT_ENGINE0 reactions · 3 comments
- Aug 4, 2026Add official Remotion and Claude productivity skills — emaddawood1991-tech/emaddawood1991-tech0 reactions
- Jul 2, 2026[capture][skills] AI-tooling watchlist from Telegram saves (Mar–Jun 2026) — tonykoop/claude-skills0 reactions
- Jun 30, 2026Follow-up: Add optional task-observer meta-skill — shakacode/agent-workflows0 reactions
- May 5, 2026Attribution to original task-observer skill — alkofu/ai-tpk0 reactions
- Apr 30, 2026[Resource]: One skill to rule them all — hesreallyhim/awesome-claude-code0 reactions · 1 comment
- Apr 29, 2026Add task observer skill to the individual skills table — travisvn/awesome-claude-skills0 reactions
- Apr 29, 2026feat(skills): add task-observer meta-skill and review-observations command — alkofu/ai-tpk0 reactions
On Hacker News · 1
- Feb 16, 2026Show HN: A Claude meta-skill that improves all your skills, including itself1 point · 1 comment
Used with
Projects with it also use ui-ux-pro-max.
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.
StrayDragonSkillui-ux-pro-max
UI/UX design intelligence. 50 styles, 21 palettes, 50 font pairings, 20 charts, 8 stacks (React, Next.js, Vue, Svelte, SwiftUI, React Native, Flutter, Tailwind). Actions: plan, build, create, design, implement, review, fix, improve, optimize, enhance, refactor, check UI/UX code. Projects: website, landing page, dashboard, admin panel, e-commerce, SaaS, portfolio, blog, mobile app, .html, .tsx, .vue, .svelte. Elements: button, modal, navbar, sidebar, card, table, form, chart. Styles: glassmorphism, claymorphism, minimalism, brutalism, neumorphism, bento grid, dark mode, responsive, skeuomorphism, flat design. Topics: color palette, accessibility, animation, layout, typography, font pairing, spacing, hover, shadow, gradient.
Also in 3 projects that use this
Other versions11 projects use a version of this.
Word for word · 0
None found.
Edited · 5
- willy92wins/DayZ-Modding-Knowledge-Pack★ 2511% same
- jeremybrasher/grokbot-skills★ 1112% same
- anupkatuwal/claude-skills★ 119% same
- beelabstudio/ai★ 036% same
- VorElze/hermes-office-skills★ 069% same
Wrote their own version · 6
- iamneilroberts/claude-skills★ 26same name
- AthArvA-188/top-10-claude-skills★ 3same name
- Carefree-1991/claude-skills★ 1same name
- mahiunfiltered/claude-code-skills★ 1same name
- mkarami313/Claude-skill★ 0same name
- okfriansyah-moh/agentic-workflow★ 0same name
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.
The file
We split this skill into the rules it gives an agent
10 rules. 0 are common, 4 are distinctive, and 6 are very specific to this project.
0
Common
Most of the 11 edited versions we read keep these.
4
Distinctive
This author's own choices. Worth a look.
6
Very specific
Tied to their project's names, files or tools. Probably skip.
Licensed with conditions — read it at the source. We count its rules but don't quote them. Read them on GitHub.
“Common” is measured: kept by at least 6 in 10 of the 11 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.
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.

