MCP Server Builder
anthropics/skills
Guides the design and implementation of Model Context Protocol servers in TypeScript or Python, from tool naming and error messages to evaluation.
Complete the Classic technical design and obtain user confirmation.
$ npx skills add rpamis/comet --skill comet-design -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install rpamis/comet comet-design --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/rpamis/comet.git skills-src && mkdir -p .claude/skills && cp -r skills-src/assets/skills/comet-design .claude/skills/comet-design && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "comet-design" agent skill from https://github.com/rpamis/comet/tree/master/assets/skills/comet-design into .claude/skills/comet-design/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "comet-design", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/rpamis/comet/tree/master/assets/skills/comet-designType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add rpamis/comet --skill comet-design -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install rpamis/comet comet-design --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/rpamis/comet.git skills-src && mkdir -p .agents/skills && cp -r skills-src/assets/skills/comet-design .agents/skills/comet-design && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "comet-design" agent skill from https://github.com/rpamis/comet/tree/master/assets/skills/comet-design into .agents/skills/comet-design/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "comet-design", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add rpamis/comet --skill comet-design -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install rpamis/comet comet-design --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/rpamis/comet.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/assets/skills/comet-design .cursor/skills/comet-design && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "comet-design" agent skill from https://github.com/rpamis/comet/tree/master/assets/skills/comet-design into .cursor/skills/comet-design/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "comet-design", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/rpamis/comet.git --path assets/skills/comet-design--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add rpamis/comet --skill comet-design -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install rpamis/comet comet-design --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/rpamis/comet.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/assets/skills/comet-design .gemini/skills/comet-design && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "comet-design" agent skill from https://github.com/rpamis/comet/tree/master/assets/skills/comet-design into .gemini/skills/comet-design/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "comet-design", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install rpamis/comet comet-designInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add rpamis/comet --skill comet-design -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/rpamis/comet.git skills-src && mkdir -p .github/skills && cp -r skills-src/assets/skills/comet-design .github/skills/comet-design && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "comet-design" agent skill from https://github.com/rpamis/comet/tree/master/assets/skills/comet-design into .github/skills/comet-design/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "comet-design", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add rpamis/comet --skill comet-design -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install rpamis/comet comet-design --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/rpamis/comet.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/assets/skills/comet-design .opencode/skills/comet-design && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "comet-design" agent skill from https://github.com/rpamis/comet/tree/master/assets/skills/comet-design into .opencode/skills/comet-design/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "comet-design", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
comet-designComplete the Classic technical design and obtain user confirmation.
Comet Design is an agent skill from rpamis/comet. Complete the Classic technical design and obtain user confirmation. Use when the user invokes /comet-design or Classic Runtime enters Design.
Its SKILL.md is about 4.3k tokens, which your agent loads only when the skill is triggered. The skill folder holds 2 other files (for example `agents/openai.yaml`).
It sits in Agent Workflows. The repository describes itself as: Comet: agent skill harness for turning ideas into evaluated workflows. The licence is MIT.
3 steps, taken from the step headings in SKILL.md.
Read from SKILL.md and the folder at commit 0fd42a0. It shows what the files ask for, not the result of running them.
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
No scripts in the folder and no shell commands in SKILL.md (its code samples are bash, yaml and markdown).
From the folder's file list and the shell code blocks in SKILL.md.
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Comet Design loads about 4.3k tokens when it runs. Until then it costs about 39 tokens; SKILL.md has 1,727 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
The full file from rpamis/comet at commit 0fd42a0, republished under its MIT licence (© rpamis). 1,727 words, ~4,288 tokens.
.claude/skills/comet-design/SKILL.md (or your agent's skills folder). This skill also uses 1 other file; get the full folder from GitHub.After the entry returns layout, use comet-classic/reference/classic-layout.md to resolve each logical root. Reuse this protocol if it is already in context. Route all OpenSpec CLI calls through the adapter and use the bound <classic-*> roots for file paths; do not run an extra root show first.
Document responsibilities: proposal records goals and scope; spec defines behavior and acceptance requirements; the Design Doc records technical decisions; the plan describes implementation steps; tasks.md records completion. Extend an existing
design.mdin place instead of creating another copy of the design. Use the file referenced bydesign_docas the formal technical design, preserving a different path already recorded for an older change. Other files reference that design rather than maintaining duplicate decisions.
Use the supported Comet CLI described in comet-classic/reference/scripts.md. When resuming from any entry, first follow comet-classic/reference/context-recovery.md to check recovery state:
comet state select <change-name>
comet state check <name> design --jsonWhen the previous phase's guard already returned this phase's state, continue from that state and agent.continuation without repeating select/check; run the entry checks above only when resuming, after workspace changes, or after external state changes. After a successful check, use the returned layout, configuration, nextAction, and coordination progress summary. Avoid individual field queries or another root show. Normal entry requires only the entry check; follow context-recovery.md when resuming without context or retrieving details. Address the reported cause if validation fails.
Recovery: Check existing artifacts and confirmation records, then complete only unfinished steps. Both normal entry and recovery preserve a registered, still-valid design and inspect data.designReadiness, data.issues, and data.nextAction. Restore missing files, correct the change associated with a file, or refresh an outdated handoff without clearing design_doc. Once the user has confirmed the design, execute the returned complete-design action; it preserves completed work. If Runtime is already in Build, the command only returns that phase's entry information.
The script must generate this pack; an Agent-written summary cannot replace it.
comet handoff <change-name> design --writeThe script generates and records the pack using the context_compression snapshot in the change's .comet.yaml.
The default context_compression: off generates:
<classic-change-dir>/.comet/handoff/design-context.json
<classic-change-dir>/.comet/handoff/design-context.mdWith beta enabled through classic.context_compression: beta in project .comet/config.yaml, copied into .comet.yaml when the change is created, it generates:
<classic-change-dir>/.comet/handoff/spec-context.json
<classic-change-dir>/.comet/handoff/spec-context.mdIt also records these fields in .comet.yaml:
handoff_context: <classic-change-ref>/.comet/handoff/design-context.json
handoff_hash: <sha256>The default pack contains script-generated source excerpts and their provenance:
design-context.json: a machine-readable index with change, phase, canonical spec, source paths, and hash.design-context.md: context for Superpowers with script markers, source path, line range, sha256, and excerpts produced by fixed rules.[TRUNCATED], with a Full source path retained.The beta pack organizes specification content in a fixed structure, reducing the OpenSpec text loaded into context while retaining the requirements needed for implementation:
spec-context.json: a machine-readable index with change, phase, mode=beta, source paths, context_hash, and each file's role under files.spec-context.md: context for Superpowers that preserves delta spec files verbatim and references related artifacts by hash.When full source context is actually needed, run:
comet handoff <change-name> design --write --fullThe pack uses OpenSpec Open artifacts:
proposal.md: goals, motivation, scope, and non-goals.design.md, when present: existing technical decisions and constraints.tasks.md: initial task scope.specs/**/spec.md: capability delta specs, preserving complete paths for nested capabilities.Immediately execute: Use the Skill tool to load the Superpowers brainstorming skill. Skipping this step is prohibited.
Include this in ARGUMENTS when loading the skill:
Language: Use the Comet artifact language from entry configuration.languageAfter the skill loads, follow its guidance and use the following context:
Change: <change-name>
OpenSpec Context Pack: <classic-change-dir>/.comet/handoff/design-context.md
For context_compression: beta, use:
OpenSpec Context Pack: <classic-change-dir>/.comet/handoff/spec-context.md
OpenSpec artifacts define the confirmed requirements. Reference them during brainstorming and discuss only unresolved technical choices; do not ask again about confirmed requirements.
Read only the selected Markdown context pack by default. Runtime validates the machine JSON; read it only to diagnose index problems. When excerpts are truncated or acceptance clauses are missing, use source path/line range to read the relevant source text. Do not read the JSON, Markdown, and every source file together.
Use the pack for technical design: implementation approach, technical risks, test strategy, and boundary conditions.
Clarify any remaining gaps in goals, scope, non-goals, acceptance scenarios, or key constraints. If the information is sufficient, develop the design directly; there is no minimum number of question rounds.
Before asking questions, read comet-classic/reference/decision-point.md. Each question needs a clear decision, a recommendation with reasons based on current constraints, and the effects of each option. Prefer an available AskUserQuestion and wait for the answer. If missing facts do not support genuine options, request the missing information directly. Reuse valid confirmations; a single technical solution does not replace the formal design confirmation in Step 1c.
Do not rewrite proposal/spec. If an OpenSpec delta spec lacks acceptance scenarios, propose a Spec Patch and write it back to that delta spec, rather than creating another requirements spec in the Design Doc. Limit Spec Patches to missing acceptance scenarios, ambiguous wording, or boundary conditions; do not substantially rewrite the delta spec's structure or scope. If a larger change is needed, record the requirements issue found during design and return to brainstorming for user confirmation.
Keep Design Doc frontmatter minimal, with only:
---
comet_change: <change-name>
role: technical-design
canonical_spec: openspec
---
Match design depth to risk. Compare 2-3 options when there are real tradeoffs. When existing architecture already determines the solution, explain why rather than inventing alternatives. High-risk interfaces, migrations, security, and concurrency require failure paths and a validation strategy.
Use only brainstorming's exploration and design methods. Present related design sections together and obtain formal confirmation once in Comet Step 1c. The external Skill must not require another approval of the complete design document, automatically invoke writing-plans, switch workspaces, or begin implementation. Do not write the Design Doc before confirmation.Do not continue without loading the skill.
If Superpowers brainstorming is unavailable, stop and ask the user to install or enable Superpowers skills. Do not replace the required step with ordinary conversation.
After the skill loads, use its method to present the proposed design in the conversation:
Brainstorming produces proposals for user confirmation in Step 1c, not a formal Design Doc. Create or update the formal design and delta spec only after confirmation. Preserve existing Open-stage design.md content; record proposed changes in a checkpoint without overwriting confirmed decisions.
Keep brainstorm-summary.md updated throughout brainstorming so discussion can resume after context compression. After any clarification or design revision that adds confirmed facts, key constraints, proposals, tradeoffs, risks, test strategy, or proposed Spec Patches, update the file. Mark unconfirmed content as “pending confirmation” or “proposed.” This checkpoint records discussion progress; it is not the Design Doc and does not replace Step 1c confirmation.
After brainstorming produces a design, follow comet-classic/reference/decision-point.md and wait for the user to explicitly confirm the design. Before that confirmation, do not create the final Design Doc, set design_doc, run the design guard, or enter /comet-build.
Present the necessary summary:
Offer these three options as a single-choice question:
Continue to Step 2 only after the user approves. If the user requests changes, continue brainstorming until they confirm the revised design.
After confirmation and before creating the Design Doc, create or update the checkpoint with the user's confirmed design.
Use file tools to ensure <classic-change-dir>/.comet/handoff/ exists; do not depend on POSIX-only directory commands.
Structure of <classic-change-dir>/.comet/handoff/brainstorm-summary.md:
# Brainstorm Summary
- Change: <change-name>
- Date: <YYYY-MM-DD>
## Confirmed Technical Approach
<Summary of the approach confirmed by the user>
## Key Tradeoffs and Risks
<Main tradeoffs and risks>
## Test Strategy
<Overview of the test approach>
## Spec Patch
<Delta spec changes to write back, or "None">Context compression: Use brainstorm-summary.md to resume interrupted discussions. Defer proactive compression until the formal design, state, and handoff are saved. If compression has already occurred, load the following as needed before continuing to Step 2:
<classic-change-dir>/.comet/handoff/brainstorm-summary.md.<classic-change-dir>/.comet/handoff/design-context.md, or beta spec-context.md, and missing source passages. Machine JSON is not required reading.brainstorm-summary.md supports recovery, but do not proactively discard the design context before saving the Design Doc. Continue directly to Step 2 and compress only after the Design Doc, state, and latest handoff have been saved.
Create the Design Doc from the full brainstorming context in the main session.
Keep frontmatter minimal:
---
comet_change: <change-name>
role: technical-design
canonical_spec: openspec
---Choose one <design-doc-path>: preserve an existing design_doc; otherwise prefer <classic-change-dir>/design.md and extend the Open-stage technical decisions in place. Use docs/superpowers/specs/YYYY-MM-DD-<topic>-design.md only when an existing project convention requires a separate Superpowers document. In that case, Open's design.md keeps only the summary required by the schema and a link to the formal design, without duplicating technical detail. Match detail to risk instead of creating empty sections or repetitive alternatives.
Write any Spec Patch to the relevant specs/**/spec.md at the same time. Maintain behavioral requirements only in the spec. The formal design references the relevant capability or acceptance clauses and must not introduce another requirements specification.
After context compression: Read brainstorm-summary.md and the handoff to restore the design discussion. If the user has not confirmed the proposal, return to Step 1b/1c. If they have, continue creating the Design Doc. The checkpoint supplies recovery details, but use the full restored context when writing the design.
After explicit user confirmation and saving the formal design, use the repository-relative path from data.artifactRefs.designDoc as <design-doc-ref>. If the user confirmed another design file, use its path relative to projectRoot. File reads and writes still use the absolute <design-doc-path>. Register the design, refresh the handoff when needed, and apply the existing Guard checks in one command:
comet state complete-design <name> --design-doc "<design-doc-ref>" --jsonRefresh the handoff whenever any source content changes, including proposal, design, task meaning, delta specs, or OpenSpec metadata; checking only Spec Patches is insufficient and the design guard will reject an outdated handoff. Skip regeneration only when all source content is unchanged. Checking task completion boxes alone does not change the requirements hash. Runtime updates state automatically; do not manually edit other fields.
Consider proactive compression only after the Design Doc and state records are saved, before Build. Confirm that design_doc, the latest handoff, handoff_hash, and the design guard result are all saved so recovery can use files without losing unrecorded design decisions.
comet_change, role: technical-design, and canonical_spec: openspec.handoff_context and handoff_hash are recorded in .comet.yaml, enforced by the guard.handoff_hash matches the current OpenSpec Open artifacts, enforced by the guard.design-context.md, or beta spec-context.md, is script-generated with traceable source path, mode, and sha256 markers, enforced by the guard.spec-context.json is structurally valid and references current source files, enforced by the guard.design_doc is recorded in .comet.yaml.comet guard <change-name> design --apply. After all checks PASS, the guard advances to phase: build; this updates phase independently of auto_transition.If Step 3 successfully returned data.phase: build, the Guard has already passed and been applied; do not run it again. On failure, address data.issues, preserve completed work, and retry the same complete-design command.
Follow comet-classic/reference/context-recovery.md with phase design.
Follow comet-classic/reference/auto-transition.md and the successful result's agent.continuation without another next query. Query again only when resuming without context, external state changes, or an older response did not include the next action:
comet state next <change-name>NEXT: auto → Load the skill named by SKILL to enter the next phase.NEXT: manual → Do not load the next skill. Return control using HINT and end this invocation without another confirmation.NEXT: done → The workflow is complete.© rpamis, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
SKILL.md and 1 other file in assets/skills/comet-design of rpamis/comet.
Open the folder on GitHubat commit 0fd42a0
Comet Design next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Comet Design this skillrpamis/comet | 3.2k | — | ~4.3k | Automated safety check: Pass | MIT | |
| MCP Server Builderanthropics/skills | 180k | 63 repos | ~2.3k | Automated safety check: Pass | Apache-2.0 | |
| Hook Development for Claude Code Pluginsanthropics/claude-plugins-official | 38k | 10 repos | ~4.1k | Automated safety check: Notes | Apache-2.0 | |
| Using Superpowersfarm-fe/farm | 5.6k | 35 repos | ~1.4k | Automated safety check: Pass | MIT | |
| Executing Plans Inlineobra/superpowers | 297k | 2 repos | ~5.1k | Automated safety check: Pass | MIT | |
| Skill CreatorAzure/azqr | 795 | 89 repos | ~8.2k | Automated safety check: Pass | Apache-2.0 |
anthropics/skills
Guides the design and implementation of Model Context Protocol servers in TypeScript or Python, from tool naming and error messages to evaluation.
anthropics/claude-plugins-official
Explains how to write Claude Code plugin hooks, both prompt-based checks and bash commands, for events such as PreToolUse, Stop and SessionStart.
farm-fe/farm
A skill your agent uses when starting any conversation - establishes how to find and use skills, requiring Skill tool invocation before ANY response including clarifying questions
obra/superpowers
Has the agent carry out an implementation plan itself, task by task in the current session, keeping a ledger, proving each step with a test and ending with one whole-branch review.
Azure/azqr
Create new skills, modify and improve existing skills, and measure skill performance.
anthropics/claude-plugins-official
Explains how to write agents for Claude Code plugins: the markdown file with YAML frontmatter, trigger descriptions, model and color settings, and system prompt design.
rpamis/comet
A skill your agent uses when 用户要启动或恢复 Comet 工作流,需要根据 active change、.comet.yaml、hotfix/tweak 意图路由到对应阶段 Skill。
rpamis/comet
Comet workflow entry. An agent skill from rpamis/comet.
rpamis/comet
Comet — OpenSpec + Superpowers dual-star development workflow.
rpamis/comet
Archive and deliver a Classic change. An agent skill from rpamis/comet.
rpamis/comet
Comet Classic workflow entry. An agent skill from rpamis/comet.
rpamis/comet
A skill your agent uses when full Comet change 已完成 open 阶段但缺少 Superpowers Design Doc,或 design 阶段需要从 OpenSpec 交接包恢复。
Categories
Complete the Classic technical design and obtain user confirmation. Comet Design is an agent skill from rpamis/comet. Complete the Classic technical design and obtain user confirmation.
Comet Design fits situations like: the user invokes /comet-design; classic Runtime enters Design.
Run `npx skills add rpamis/comet --skill comet-design -a claude-code`. Or copy the skill folder (assets/skills/comet-design in rpamis/comet) into .claude/skills/comet-design in your project. Claude Code loads it when a task matches its description.
Run `npx skills add rpamis/comet --skill comet-design -a codex`. Or copy the skill folder (assets/skills/comet-design in rpamis/comet) into .agents/skills/comet-design in your project. Codex loads it when a task matches its description.
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add rpamis/comet --skill comet-design -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/comet-design, .gemini/skills/comet-design, .github/skills/comet-design and .opencode/skills/comet-design in your project.
SKILL.md names no scripts, command-line tools or credentials: Comet Design is instructions for the agent only.
SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.
Comet Design is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
About 4.3k tokens (SKILL.md is roughly 17k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.
Skills that share tags, products or a category with Comet Design: MCP Server Builder (anthropics/skills, 180k stars), Hook Development for Claude Code Plugins (anthropics/claude-plugins-official, 38k stars), Using Superpowers (farm-fe/farm, 5.6k stars) and Executing Plans Inline (obra/superpowers, 297k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
rpamis (a GitHub organization) maintains it in rpamis/comet, which has 3,166 GitHub stars. The repository holds 52 skills in this directory. The repository was last updated on October 9, 2026.
Source: rpamis/comet on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.