task SKILL.md skillB

task from vzakharov/muthur is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.

Install it, pinned
mdr add vzakharov/muthur/task@git:20261014.0f16275

Pins git:20261014.0f16275 in mdr.lock. Pin the fingerprint instead.

Audit grade
B, 16 of 17 checks passed
Kind
SKILL.md skill
Repository
vzakharov/muthur
Path
.claude/skills/task/SKILL.md
Versions recorded
5
Latest version
git:20261014.0f16275, committed Oct 14, 2026
First version recorded
Sep 23, 2026
Size
7,031 bytes
Repository stars
22
License
MIT

Take a task and decide for yourself whether it needs a plan, write the plan when it does, and then decide whether the operator has to look — plan and hand off, plan and then implement, or implement with no plan at all. Invoke as `/task <what to do>`. The prompt is a conditional go-ahead scoped to that one task.

Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.

What the file says

The decision this skill makes, before any other: **does this task get a plan,
and does the plan block on the operator?** Those are two questions with a plan
file between them: the second judges the plan, so it is asked once the plan
exists.

**The prompt is a conditional go-ahead.** It authorizes implementation _on
condition that_ the agent judges the operator's gate unnecessary, scoped to the
task in that message and to that message alone.

`/task <what to do>` is the invocation, and it is also where an unrouted
session's first change lands: CLAUDE.md § "Plan mode & questions in web
sessions" routes here the first turn that asks for a change to this codebase,
whether that is the opening prompt or a follow-up to one that was answered, so
most tasks arrive without anyone typing the name. A task routed mid-session is
the work as the conversation now stands. A `#<N>` in that prompt means the
thread is exported and committed before anything else happens — `/take-issue`
does that and hands the number back. Both mentions stay bare rather than
`@`-references: this skill ships to adopters who track no issues at all, and
nothing here needs those files read.
…

Read the whole file at its exact version.

How to install

The exact contents, by fingerprint
mdr add vzakharov/muthur/task@sha256:cb2c737236c1e822

The line at the top of this page pins the label git:20261014.0f16275, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.

Badge

mdr badge

[![mdr](https://markdownregistry.com/badge/art_bxw4a2s5i64eblta.svg)](https://markdownregistry.com/a/art_bxw4a2s5i64eblta)

1 badge image fetches in 30 days. What the badge shows.

Versions

The registry has recorded 5 versions of this file, between Sep 23, 2026 and Oct 14, 2026, about one every 5 days.

versioncommittedcommitsizeaudit
git:20261014.0f16275 latest2026-10-14 0f16275 7,031 BB view · diff
git:20260930.3322c8f2026-09-30 3322c8f 7,009 BB view · diff
git:20260925.c493c242026-09-25 c493c24 6,959 BB view · diff
git:20260924.634bdfa2026-09-24 634bdfa 6,825 BB view · diff
git:20260923.d6730fb2026-09-23 d6730fb 6,676 BB view

Audit of the latest version

B  16 of 17 checks passed. Fixed automatic checks with no AI involved, so the same checks give the same answer every run.
  • fail: Frontmatter declares a name
  • pass: Frontmatter block present
  • pass: Frontmatter declares a description
  • pass: Size between 200 bytes and 200 KB (7031 bytes)
  • pass: No zero-width or bidi control characters
  • pass: No instruction hidden inside an HTML comment
  • pass: No link to an exfiltration or paste host
  • pass: No credential-shaped string
  • pass: No instruction to send local credentials anywhere
  • pass: No text hidden with inline styles
  • pass: No prompt-injection phrasing
  • pass: No curl or wget piped into a shell
  • pass: No recursive delete of root, home or parent
  • pass: No instruction to read or print local credentials
  • pass: No base64 blob over 200 characters
  • pass: No link to a raw IP address
  • pass: No script tag

Starting a new one? Use a template for SKILL.md on mdpantry, which passes every one of these checks and says what to change in it.

Source

GitHub

vzakharov/muthur · 22 stars · license MIT · pushed 2026-10-30 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_bxw4a2s5i64eblta
GET https://markdownregistry.com/api/v1/resolve?ref=vzakharov/muthur/task
GET https://markdownregistry.com/api/v1/blob/cb2c737236c1e82273121bfaf5d20b4585e2ad02d30d7c0756c39cfe7b414c07

Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.

Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.

More from vzakharov/muthur

CLAUDE.md@.claude/cold-cache CLAUDE.md file
vzakharov/muthur · .claude/cold-cache/CLAUDE.md
git:20261026.79f8fb3 · audit A · 22 stars
CLAUDE.md@.claude/context-budget CLAUDE.md file
vzakharov/muthur · .claude/context-budget/CLAUDE.md
git:20261029.ade5154 · audit A · 22 stars
CLAUDE.md@.claude/costs CLAUDE.md file
vzakharov/muthur · .claude/costs/CLAUDE.md
git:20261029.ade5154 · audit A · 22 stars
CLAUDE.md@.claude/keepalive CLAUDE.md file
vzakharov/muthur · .claude/keepalive/CLAUDE.md
git:20261027.34a00c5 · audit A · 22 stars
CLAUDE.md@.claude/skills CLAUDE.md file
vzakharov/muthur · .claude/skills/CLAUDE.md
git:20261014.0f16275 · audit A · 22 stars
audit-github-backlog SKILL.md skill
vzakharov/muthur · .claude/skills/audit-github-backlog/SKILL.md · Sweep the whole open GitHub backlog — every open issue and PR — judging each against today's code, and produce a…
git:20261014.0f16275 · audit B · 22 stars
bootstrap-workflow-dispatch SKILL.md skill
vzakharov/muthur · .claude/skills/bootstrap-workflow-dispatch/SKILL.md · Temporarily bootstrap a newly renamed or newly added `workflow_dispatch` workflow so GitHub can dispatch it before the…
git:20261014.0f16275 · audit B · 22 stars
branch-rename SKILL.md skill
vzakharov/muthur · .claude/skills/branch-rename/SKILL.md · Rename the current auto-generated session branch (e.g. `claude/<adjective>-<noun>-<hash>`) to a semantic name derived…
git:20261014.0f16275 · audit B · 22 stars
check-merge SKILL.md skill
vzakharov/muthur · .claude/skills/check-merge/SKILL.md · Check once whether the branch's merge target (the branch its PR is based on) advanced or the PR landed since the branch…
git:20261027.34a00c5 · audit B · 22 stars
detemplate SKILL.md skill
vzakharov/muthur · .claude/skills/detemplate/SKILL.md · Turn a fresh "Use this template" fork into an actual project: profile the tree, decide group by group what travels, and…
git:20261026.67c8338 · audit B · 22 stars
dry SKILL.md skill
vzakharov/muthur · .claude/skills/dry/SKILL.md · Review code from the current (or clearly-restored) session for DRY opportunities. Apply the obvious wins immediately…
git:20261014.0f16275 · audit B · 22 stars
finalize SKILL.md skill
vzakharov/muthur · .claude/skills/finalize/SKILL.md · Finalize (a.k.a. "prep merge") — land prep: verify there's a draft PR, run the quality passes (`/polish`), run the vet…
git:20261029.b2079f0 · audit B · 22 stars

Every file in vzakharov/muthur

Other files named task

task SKILL.md skill
codeaholicguy/ai-devkit · skills/task/SKILL.md · AI DevKit · Track dev-lifecycle / dev-debug progress on a durable task with the ai-devkit task CLI. Use to record…
git:20261023.29155fd · audit A · 1,656 stars
task SKILL.md skill
mweinbach/agent-coworker · skills/task/SKILL.md · Hand ordinary chat off to durable Task mode when the user explicitly invokes /task or requests a managed task…
git:20260905.cfba839 · audit A · 157 stars
task SKILL.md skill
uipath/coder_eval · plugins/coder-eval/skills/task/SKILL.md · Turn a natural-language description into coder-eval task YAML — minimal prompts, weighted criteria that check output…
git:20260924.28c846e · audit B · 156 stars
task SKILL.md skill
gtapps/hermitd · plugins/hermitd/skills/task/SKILL.md · Tracks chat assignments, steering in a task thread, WORKER completions, and confirmations alongside task progress and…
git:20261025.bad50e7 · audit A · 83 stars
task SKILL.md skill
gtapps/claude-code-hermit · plugins/hermitd/skills/task/SKILL.md · Tracks chat assignments, steering in a task thread, WORKER completions, and confirmations alongside task progress and…
git:20261031.93f727d · audit A · 83 stars
task SKILL.md skill
theafh/ai-modules · plugins/ai_dev/skills/task/SKILL.md · Manage the project task backlog as plain markdown files in tasks/. Use when the user asks to list or query existing…
v1.3.50 · audit B · 39 stars
task SKILL.md skill
camoa/claude-skills · aida/skills/task/SKILL.md · This skill should be used when the user wants to "create a task", "start a new task", "split a task", "make this an…
git:20261028.aaa1221 · audit A · 33 stars
task SKILL.md skill
ulises-jeremias/agent-toolkit · plugins/agent-toolkit-complete/.opencode/skills/task/SKILL.md · WHAT - Draft and review technical tasks using the Best Practices Task Template; includes
git:20260831.856ee57 · audit A · 19 stars

Browse by kind, by grade B, or by owner.