Agent skill

Comet Hotfix

by rpamis in rpamis/comet

Use the Classic preset to repair a localized defect. An agent skill from rpamis/comet.

MITAuto-check passedAgent Workflows

Install Comet Hotfix

skills CLI
$ npx skills add rpamis/comet --skill comet-hotfix -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install rpamis/comet comet-hotfix --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/rpamis/comet.git skills-src && mkdir -p .claude/skills && cp -r skills-src/assets/skills/comet-hotfix .claude/skills/comet-hotfix && rm -rf skills-src

Use ~/.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/

Facts

Skill name
comet-hotfix
GitHub stars
3.2k
Token cost
~5.1k tokens
SKILL.md length
2,631 words
Files
2
Skills in repo
52
Repo updated
First seen
Licence
MIT

At a glance

Use the Classic preset to repair a localized defect. An agent skill from rpamis/comet.

  • Works in 6 steps: Set the output language → Open a minimal change → Implement directly → …
  • The user explicitly invokes /comet-hotfix
  • SKILL.md covers Preset flow: 6 steps, Continue through the preset, Escalation decisions and Exit conditions, plus 1 more section
  • Calls mvn and npm

What it does

Comet Hotfix is an agent skill from rpamis/comet. Use the Classic preset to repair a localized defect. Use when the user explicitly invokes /comet-hotfix, selects hotfix, or resumes workflow: hotfix.

Its SKILL.md is about 5.1k 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.

When your agent uses it

  • The user explicitly invokes /comet-hotfix
  • Resumes workflow: hotfix

Example prompts

  • “/comet-hotfix”

Workflow steps

6 steps, taken from the step headings in SKILL.md.

  1. Set the output language
  2. Open a minimal change
  3. Implement directly
  4. Confirm the root cause is eliminated
  5. Verify
  6. Archive

What it can do on your machine

Read from SKILL.md and the folder at commit 0fd42a0. It shows what the files ask for, not the result of running them.

  • Tool permissions

    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.

  • Runs code

    Shell commands in SKILL.md call:

    • mvn
    • npm

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md. Its commands use npm, which can reach the network depending on how they are called.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Comet Hotfix loads about 5.1k tokens when it runs. Until then it costs about 41 tokens; SKILL.md has 2,631 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~41
When it runs · the whole SKILL.md, loaded when a task matches
~5.1k

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.

Safety

Auto-check passed

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.

SKILL.md

The full file from rpamis/comet at commit 0fd42a0, republished under its MIT licence (© rpamis). 2,631 words, ~5,126 tokens.

Download SKILL.mdSave it as .claude/skills/comet-hotfix/SKILL.md (or your agent's skills folder). This skill also uses 1 other file; get the full folder from GitHub.
name
comet-hotfix
description
Use the Classic preset to repair a localized defect. Use when the user explicitly invokes /comet-hotfix, selects hotfix, or resumes workflow: hotfix.

Comet Preset: Hotfix

Before starting or resuming, read and follow comet-classic/reference/classic-layout.md. All OpenSpec CLI calls must use the adapter, and all paths must use the bound <classic-*> logical roots.

A short defect-repair flow: open → build → root cause check → verify → archive. It skips brainstorming and a full implementation plan, and applies to repairing existing behavior without designing new features.

All applicability conditions must hold:

  1. Repair an existing feature's defect without adding a feature.
  2. No interface changes or architectural redesign.
  3. The scope can be estimated. File counts are only a prompt for review, not an automatic escalation rule; see “Escalation decisions.”

When the preset may no longer fit: If the repair encounters changes listed under “Escalation decisions,” let the user decide whether to use the full /comet-classic flow.


Preset flow: 6 steps

0. Set the output language

Use Comet's configured artifact language for the reduced OpenSpec artifacts. Before .comet.yaml exists, read classic.language from project .comet/config.yaml, then global ~/.comet/config.yaml. After initialization, read it with comet state get <name> language.

Execution order: open → build → root cause check → verify → archive. Hotfix presets how each stage runs: prepare the necessary artifacts, implement directly, check that the root cause is eliminated, choose verification based on size, then request final archive confirmation after verification passes.

Use the supported Comet CLI described in comet-classic/reference/scripts.md. On recovery from any entry, first check phase/workflow under comet-classic/reference/context-recovery.md.

For an existing hotfix, the first state operation must be comet state select <change-name>. For a new change, run it immediately after .comet.yaml initializes successfully and before source edits.

After entering the hotfix workspace and reading current phase, run comet task <project-root> --task "<original-user-request>" --phase "<phase>" --session "<stable-task-session-id>" --json. Use the returned context as follows:

  • Add only returned text to the current context. Context Manifest (manifest / <context_manifest>) contains only summaries, application reasons, and stable IDs. Add --expand-context "<id>" when source text, provenance, or validation details are needed. When path, operation, or phase changes, select applicable entries again using the same --session.
  • Use comet memory remember ... --scope global|project when the user explicitly asks for long-term memory. Use comet memory observe only for implicit, reusable, stable collaboration patterns. Do not save task summaries, progress, command output, or test results.
  • Before task completion, write project-validated and reusable experience to Project Memory with comet knowledge remember <project-root> --title "<concise title>" --text "<symptom, approach, verification>" --type <fact|decision|pattern|procedure|constraint|failure-resolution> --json. The same title updates the existing entry; skip the write when nothing is reusable, and never save task summaries, one-off command output, or unverified guesses in Project Memory.
  • After actually using an entry and determining its outcome, take applications[].applicationId (application_id in Hook text) and run comet task <project-root> --task "<original-user-request>" --application "<application-id>" --outcome used-successfully|ignored|overridden|corrected|contributed-to-failure --json to record the result.
  • Before task completion, complete one learning check: when a user correction, preference, or collaboration habit has a clear reuse condition, call comet memory observe and pass --learning-check submitted to the completion command; pass --learning-check no-observation when you checked and found none, or --learning-check not-run when no check occurred. Use observation JSON learning.result and status.learning.lastCheck to distinguish candidate, promotion, deduplication, and skip; do not submit task summaries or test results.
  • At task completion, still run comet task with --complete --workflow <workflow> --change <change-id> --learning-check submitted|no-observation|not-run. Without Hooks, this Skill uses the same interface. comet memory context is a compatibility entry only. Plugin failures do not block the repair.
1. Open a minimal change

Reuse Comet Open with hotfix defaults. Skip the full openspec-explore exploration and create only the artifacts needed for the repair.

Required now: Load openspec-new-change using the Skill tool. Do not skip this step.

<!-- external-openspec-skill-override -->

Adapt external OpenSpec instructions: Do not directly invoke the official CLI, adopt a fixed cwd, or read/write fixed physical OpenSpec paths. Use comet classic openspec -- <args...> for every OpenSpec command and this invocation's <classic-*> logical roots for all change and artifact paths.

Workspace isolation is a user choice made before state initialization; do not write current as an assumed default (choosing a worktree after initialization fails: the change directory only appears in the primary root, not in the worktree). Pause under comet-classic/reference/decision-point.md and present:

  • A. Work on the current branch (--isolation current, binding the actual branch).
  • B. Create a branch: create and switch to hotfix/YYYYMMDD/<change-name> (--isolation branch).
  • C. Create a worktree: first load Superpowers using-git-worktrees with the Skill tool and let it create the isolated workspace (--isolation worktree).

Then prepare the workspace and initialize plus select the change inside the returned projectRoot (resume follows the same order):

bash
comet classic workspace prepare <name> --isolation <selected-isolation> --json
cd <returned projectRoot>
comet state init <name> hotfix --isolation <selected-isolation>
comet state select <name>
comet state check <name> open

If select/check returns BLOCKED — or a branch-binding ERROR — because bound_branch differs from the current branch, pause under comet-classic/reference/decision-point.md. Offer a single choice: return to the bound branch and rerun entry checks, or, after the user explicitly confirms that the current branch should take over this change, run comet state rebind <change-name> and rerun entry checks. Do not switch or rebind branches yourself.

Then create the reduced artifacts:

  • proposal.md: problem, root cause, and repair goal; no solution comparison required.
  • design.md: the repair approach; one approach is enough.
  • tasks.md: repair tasks.
  • No delta spec is required unless the fix changes acceptance scenarios in an existing spec.

Apply the phase guard to move from open to build:

bash
comet guard <change-name> open --apply

Check auto_transition to decide whether to continue:

bash
comet state next <name>
  • NEXT: auto: continue to Step 2.
  • NEXT: manual: follow HINT, return control, and end this invocation. Do not ask for another continuation approval.
2. Implement directly

Use hotfix defaults: build_mode: direct, tdd_mode: direct, review_mode: off. Preserve the isolation confirmed in Step 1; do not change it back to current.

direct skips full planning and per-task TDD orchestration; it still requires reproduction, regression tests, and verification. Skip Superpowers brainstorming and writing-plans. Task count alone does not trigger /comet-build. Execute even a longer tasks.md in order within the current hotfix. Ask whether to escalate to full only when a later escalation condition applies, or when the file-count threshold is exceeded without valid authorization.

Before starting or resuming edits, handle uncommitted changes under comet-classic/reference/dirty-worktree.md. After establishing ownership, apply “Escalation decisions” if the repair meets an escalation condition or exceeds the file-count prompt.

Before changing implementation, reproduce the issue and record the failure:

  1. Confirm the reported old behavior actually fails with minimal repeatable steps. Record the command, input, and actual result.
  2. Where automation is possible, add and run a failing regression test first. Confirm that it fails for this defect, not an environment or test error.
  3. Where automation is not currently possible, record why and provide repeatable manual failure evidence in the proposal/verification report. Do not edit code without failure evidence.

After obtaining RED evidence, execute tasks.md in order:

  1. Read <classic-change-dir>/tasks.md for unfinished tasks.
  2. For each task:
    • Implement the described repair.
    • Run the project's formatter, such as mvn spotless:apply or npm run format.
    • Run the new regression test until it passes, then run relevant tests.
    • Change its tasks.md - [ ] to - [x].
    • Commit using fix: <repair-summary>.
  3. Explicitly run the relevant project tests and build after all tasks are complete.

During hotfix, a crash, unexpected behavior, failing test, or failing build encountered while running the program, tests, build, or manual verification requires loading Superpowers systematic-debugging through the Skill tool. Do not propose or implement source repairs before completing root-cause investigation.

Follow comet-classic/reference/debug-gate.md for investigation, the minimal failing test, verification after repair, and completing those steps in the current change.

If the fix affects existing spec acceptance scenarios:

  • Create <classic-change-dir>/specs/<capability>/spec.md as a delta spec.
  • Include only ## MODIFIED Requirements.
3. Confirm the root cause is eliminated

Do this before the build guard to confirm that the repair actually removes the cause:

  1. Read the bug description and root cause in proposal.md.
  2. Search the relevant code and confirm the faulty implementation has been removed or corrected.
  3. If the cause remains, return to Step 2. Phase is still build, so no state rollback is needed.

Escalation prompts:

  • The check reveals a deeper architecture issue: pause under “Escalation decisions” and let the user choose whether to use the full flow.
  • The repair needs another interface change, such as a new public API: pause under the same section for the user's decision.

After confirming elimination, advance from build to verify:

bash
comet guard <change-name> build --apply

State becomes phase: verify, verify_result: pending; continue to verification.

4. Verify

Reuse /comet-verify, whose size assessment chooses light or full verification.

Required now: Load comet-verify using the Skill tool. Do not skip this step.

A small hotfix without delta spec usually meets light conditions (≤ 3 tasks and changed files below the scale threshold). Follow comet-verify's light-verification checklist. Default review_mode: off does not dispatch automatic code review. If the user wants review, they can set comet state set <name> review_mode standard or thorough before verification. If the hotfix creates delta spec, follow comet-verify's scale rules into full verification.

After verification and its report are complete, /comet-verify runs comet guard <change-name> verify --apply. Reuse its successful archive state and agent.continuation; do not manually set verify_result: pass or repeat Guard. If Guard fails, address its reasons without claiming acceptance passed. /comet-archive still requires final confirmation before archiving; never run archive automatically.

5. Archive

Reuse /comet-archive. Require .comet.yaml verify_result: pass and wait for its final archive confirmation.

Required now: Load comet-archive using the Skill tool. Do not skip this step.

If there is delta spec, sync it to main spec under comet-archive rules and apply archive annotations to the linked Design Doc and Plan.


Show full SKILL.md (1,083 more words)Show less

Continue through the preset

<IMPORTANT>
Hotfix runs continuously by default. After `/comet-hotfix`, automatically move through its own steps without extra pauses. If `auto_transition: false`, end the invocation between build/verify/archive phases and use `HINT` to tell the user how to invoke the next phase later. Do not add another confirmation question. Regardless of auto_transition, pause for these user decisions:
  1. An escalation condition appears: pause, present choices, and wait for an explicit decision to continue hotfix or move to the full /comet-classic flow.
  2. Verify needs acceptance of a WARNING/SUGGESTION deviation, a spec-divergence decision, or a strategy after the automatic repair limit. The first 3 clearly repairable failures are repaired and reverified automatically.
  3. The final pre-archive choice of whether to archive and how to deliver the archive commit.

Order: quick Open → direct Build → root-cause elimination check → Verify → Archive → done.

Continue to the next phase as soon as the current one finishes, subject to the rules above. Still invoke the required Comet/OpenSpec/Superpowers skills within each phase. If a called skill has a user decision, follow its rules. </IMPORTANT>


Escalation decisions

Escalation decides only whether to replace the preset with full. File count does not automatically upgrade the workflow. comet state scale recommends light/full verification without writing configuration; Verify chooses based on actual risk.

If /comet-classic passes an intent frame, before Build recheck only risk_signal and whether work adds a feature or public API, changes a structured-data schema, needs cross-module coordination, or exposes a deeper architecture issue. Follow this section when these arise; do not repeat entry intent classification.

During repair, watch for:

  • Coordinated edits across modules.
  • A new feature.
  • Database schema changes.
  • A new public API.
  • A deeper architecture issue, often discovered by the root-cause elimination check.

For any of these, the Agent must neither escalate nor decide to stay on hotfix without the user.

File count prompts a scope review only; it is not a substantive escalation signal. When delivery files exceed the prompt threshold, such as > 4 files, first count the current change's delivery files and check for valid authorization below. More files do not necessarily require the full flow. Defect repairs usually involve 1–3 files; exceeding the threshold warrants checking whether the preset still fits.

Delivery files include only implementation/source, tests, user documentation, configuration, and generated output. Count committed changes after the confirmed baseline together with staged, unstaged, and untracked files, deduplicated by path. Exclude OpenSpec artifacts in the current change directory, .comet metadata, and unrelated dirty files. Do not count OpenSpec artifacts or unrelated dirty files toward the threshold, and do not skip substantive-signal checks because the file count is small.

Reusing explicit authorization for a file-count-only prompt

Only an explicit authorization from the user of the current change to continue when the scope and risk are unchanged and file count is the only trigger can create or reuse authorization. Ordinary “start repairing” instructions, Skill invocation, historical preferences, and Personal Memory are not authorization. Store it only in <classic-change-dir>/.comet/rulings.md using this stable structure; if the change directory does not exist yet, keep the record pending and do not assume authorization exists:

text
### Preset file-count authorization
- status: active|invalidated
- workflow: hotfix
- decision: continue-on-file-count-only
- scope: <confirmed scope of the current change>
- allowed-file-categories: implementation, tests, user-docs, config, generated
- authorization-basis: user-explicit
- reason: <basis and reason supplied by the user>

When the file-count threshold is first exceeded without a valid ruling, pause under comet-classic/reference/decision-point.md and ask the user to choose continue hotfix (A) or escalate to full (B). Write the ruling only after the user explicitly chooses A and confirms that scope and risk are unchanged and file count is the only trigger; later growth within the same scope can reuse status: active without asking again. With valid authorization, do not ask, but first report the total file count, category breakdown, mapping to the confirmed scope, the ruling location, and the evidence that no substantive escalation signal was found, then continue hotfix. Verification depth still follows actual risk and the comet state scale recommendation.

When resuming a task, read the current change's ruling first. A valid status: active record is reused under the same conditions; status: invalidated, missing, unreadable, ambiguous, or scope-mismatched records require pausing and offering continue hotfix or escalate to full again.

status: active is valid only when workflow matches hotfix, scope, acceptance, and risk assumptions are unchanged, all new delivery files remain within allowed-file-categories, and file count is the only trigger. If the user revokes authorization, the workflow changes, scope or acceptance changes, a new module/API/schema/capability/architecture issue appears (for example, a new public API or structured-data schema change), or a file falls outside the authorized categories, mark the record status: invalidated and return to the pause. If rulings.md is missing, unreadable, ambiguous, or invalid, there is no valid authorization and the Agent must pause; when escalating to full, also mark it status: invalidated and do not reuse it.

Therefore, any substantive escalation signal, or a file-count threshold exceeded without valid authorization, requires pausing under comet-classic/reference/decision-point.md and waiting for an explicit choice. Do not enter /comet-design or create a Design Doc automatically.

After the user chooses escalation (B), run the supported state-machine transition to full and return to design:

bash
comet state transition <name> preset-escalate

It atomically sets workflow/classic_profile to full, moves phase to design, clears design_doc, and clears preset-specific build_mode, tdd_mode, review_mode, isolation, and verify_mode, together with workspace bindings such as bound_branch — before entering Build, re-decide isolation and rebind (state set <name> isolation ...) under comet-classic/reference/decision-point.md, or guard will reject the missing isolation. Immediately load comet-design using the Skill tool to complete the design within the existing change. On entering Build, jointly reconfirm the complete working configuration.

If the user chooses to continue (A), continue hotfix only after they explicitly confirm that scope and risk are unchanged and file count is the only trigger, and record the authorization basis and reason in the structure above. A bare “start repairing” or “continue” without that authorization meaning is not sufficient; pause for clarification.


Exit conditions

  • The defect is fixed and tests pass.
  • The change is archived.
  • Any spec changes are synced to main spec.
  • Phase guards: use comet guard <change-name> build --apply before build → verify, and follow /comet-verify to run comet guard <change-name> verify --apply before verify → archive.

Continue to the next phase

Follow comet-classic/reference/auto-transition.md and agent.continuation from the successful result. Do not repeat next, select, or check while valid state information is available. Run this only after context loss, external state changes, or when an older result lacks that information:

bash
comet state next <name>
  • NEXT: auto: invoke the skill named by SKILL: build returns comet-hotfix, verify returns comet-verify, and archive returns comet-archive.
  • NEXT: manual: do not invoke the next skill. Follow HINT, return control, and end this invocation without another confirmation question.
  • 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

Files

SKILL.md and 1 other file in assets/skills/comet-hotfix of rpamis/comet.

  • SKILL.md
  • agents/openai.yaml

Open the folder on GitHubat commit 0fd42a0

Compare with similar skills

Comet Hotfix 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.

Comet Hotfix compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Comet Hotfix this skillrpamis/comet3.2k—~5.1kAutomated safety check: PassMIT
MCP Server Builderanthropics/skills180k63 repos~2.3kAutomated safety check: PassApache-2.0
Hook Development for Claude Code Pluginsanthropics/claude-plugins-official38k10 repos~4.1kAutomated safety check: NotesApache-2.0
Using Superpowersfarm-fe/farm5.6k35 repos~1.4kAutomated safety check: PassMIT
Executing Plans Inlineobra/superpowers297k2 repos~5.1kAutomated safety check: PassMIT
Skill CreatorAzure/azqr79589 repos~8.2kAutomated safety check: PassApache-2.0

Similar skills

  • MCP Server Builder

    anthropics/skills

    Official

    Guides the design and implementation of Model Context Protocol servers in TypeScript or Python, from tool naming and error messages to evaluation.

    180k GitHub starsUsed in 63 repos~2.3k tokens
    Agent WorkflowsAuto-check passed
  • Hook Development for Claude Code Plugins

    anthropics/claude-plugins-official

    Official

    Explains how to write Claude Code plugin hooks, both prompt-based checks and bash commands, for events such as PreToolUse, Stop and SessionStart.

    38k GitHub starsUsed in 10 repos~4.1k tokens
    Agent WorkflowsAuto-check: notes
  • Using Superpowers

    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

    5.6k GitHub starsUsed in 35 repos~1.4k tokens
    Agent WorkflowsAuto-check passed
  • Executing Plans Inline

    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.

    297k GitHub starsUsed in 2 repos~5.1k tokens
    Agent WorkflowsAuto-check passed
  • Skill Creator

    Azure/azqr

    Official

    Create new skills, modify and improve existing skills, and measure skill performance.

    795 GitHub starsUsed in 89 repos~8.2k tokens
    Agent WorkflowsAuto-check passed
  • Claude Code Agent Development

    anthropics/claude-plugins-official

    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.

    38k GitHub starsUsed in 7 repos~2.8k tokens
    Agent WorkflowsAuto-check passed

More from rpamis/comet

All 52 skills in this repo
  • Comet

    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。

    3.2k GitHub stars~2.7k tokensUpdated today
    Auto-check passed
  • Comet

    rpamis/comet

    Comet workflow entry. An agent skill from rpamis/comet.

    3.2k GitHub stars~686 tokensUpdated today
    Auto-check passed
  • Comet

    rpamis/comet

    Comet — OpenSpec + Superpowers dual-star development workflow.

    3.2k GitHub stars~4.4k tokensUpdated today
    Auto-check passed
  • Comet Archive

    rpamis/comet

    Archive and deliver a Classic change. An agent skill from rpamis/comet.

    3.2k GitHub stars~3.9k tokensUpdated today
    Auto-check passed
  • Comet Classic

    rpamis/comet

    Comet Classic workflow entry. An agent skill from rpamis/comet.

    3.2k GitHub stars~3k tokensUpdated today
    Auto-check passed
  • Comet Design

    rpamis/comet

    Complete the Classic technical design and obtain user confirmation.

    3.2k GitHub stars~4.3k tokensUpdated today
    Auto-check passed

Categories

Questions about Comet Hotfix

What does Comet Hotfix do?

Use the Classic preset to repair a localized defect. An agent skill from rpamis/comet. Comet Hotfix is an agent skill from rpamis/comet. Use the Classic preset to repair a localized defect.

When should I use Comet Hotfix?

Comet Hotfix fits situations like: the user explicitly invokes /comet-hotfix; resumes workflow: hotfix.

How do I install Comet Hotfix in Claude Code?

Run `npx skills add rpamis/comet --skill comet-hotfix -a claude-code`. Or copy the skill folder (assets/skills/comet-hotfix in rpamis/comet) into .claude/skills/comet-hotfix in your project. Claude Code loads it when a task matches its description.

How do I install Comet Hotfix in Codex?

Run `npx skills add rpamis/comet --skill comet-hotfix -a codex`. Or copy the skill folder (assets/skills/comet-hotfix in rpamis/comet) into .agents/skills/comet-hotfix in your project. Codex loads it when a task matches its description.

Can I use Comet Hotfix in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add rpamis/comet --skill comet-hotfix -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-hotfix, .gemini/skills/comet-hotfix, .github/skills/comet-hotfix and .opencode/skills/comet-hotfix in your project.

What does Comet Hotfix need to run?

Going by SKILL.md and its folder, Comet Hotfix needs the command-line tools its instructions call (mvn and npm).

Does Comet Hotfix access the network?

SKILL.md contains no URLs. Its commands use npm, which can reach the network depending on how they are called. This is read from the text; nothing was executed.

Is Comet Hotfix safe to install?

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.

What licence does Comet Hotfix use?

Comet Hotfix is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Comet Hotfix use?

About 5.1k tokens (SKILL.md is roughly 21k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Comet Hotfix?

Skills that share tags, products or a category with Comet Hotfix: 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.

Who maintains Comet Hotfix?

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.