New: 98,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

Prismer Skill Creator

by Prismer-AI in Prismer-AI/PrismerCloud

Walks an agent through creating, importing, editing, validating, testing and publishing Prismer Skills with a fixed workflow and bundled scripts.

MITAuto-check: notesAgent Workflows

Install Prismer Skill Creator

skills CLI
$ npx skills add Prismer-AI/PrismerCloud --skill skill-creator -a claude-code

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

GitHub CLI
$ gh skill install Prismer-AI/PrismerCloud skill-creator --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/Prismer-AI/PrismerCloud.git skills-src && mkdir -p .claude/skills && cp -r skills-src/sdk/cloud/catalog/skills/skill-creator .claude/skills/skill-creator && 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
skill-creator
GitHub stars
1.6k
Token cost
~2.6k tokens
SKILL.md length
983 words
Files
31 (incl. scripts, references, assets)
Skills in repo
88
Repo updated
First seen
Licence
MIT

At a glance

Walks an agent through creating, importing, editing, validating, testing and publishing Prismer Skills with a fixed workflow and bundled scripts.

  • Works in 7 steps: Validate required input, types, target,… → Reject secret-bearing flags, including… → Read credentials and remote-write… → …
  • Turning a repeatable capability into a Prismer skill
  • SKILL.md covers Ownership boundary, Single bundle, Fixed authoring closure and External Skill libraries, plus 3 more sections
  • Runs Python scripts from its folder; calls node; needs SERVICE_API_KEY

What it does

The skill sends each request down one of four routes: a single new or edited skill, a repository or directory holding several skills (which uses a separate import guide), improving or benchmarking an existing skill, or building a role or persona template, which it hands to role-builder. It says to pick a route and act, without first probing cloud --help, creating test catalog objects or calling raw skill endpoints.

Repeated, stable work lives in bundled scripts, while SKILL.md only routes the request, declares configuration and explains results; product resource operations belong in the service, SDK or CLI. The package ships evaluation helpers, including analyzer, comparator and grader agent prompts and a local review viewer. Optional paid trigger evaluations need explicit budget approval and the PRISMER_ALLOW_PAID_EVAL flag, run in isolated temporary skill directories, and treat an unsupported CLI, timeout or auth failure as an error rather than a negative result.

When your agent uses it

  • Turning a repeatable capability into a Prismer skill
  • Importing a repository of skills into a workspace or Marketplace
  • Improving or benchmarking an existing Prismer skill
  • Validating a skill bundle before publishing it to Studio

Example prompts

  • “Turn our release checklist into a Prismer skill and validate it.”
  • “Import the skills in the ./team-skills folder into my workspace.”
  • “Improve the description of my invoice-parser skill and test whether it triggers correctly.”

Requirements

  • The Prismer SDK or cloud CLI
  • A current Claude CLI with the --bare flag for optional paid trigger evaluations
  • Compatibility (from SKILL.md): ["prismer-sdk","hermes","claude-code","codex","openclaw"]
  • Pre-approved tools (allowed-tools): Read, Write, Bash, WebFetch

Workflow steps

7 steps, taken from the first numbered list in SKILL.md.

  1. Validate required input, types, target, and environment locally.
  2. Reject secret-bearing flags, including --token=value; do not echo the
  3. Read credentials and remote-write authority only from trusted environment
  4. Run a read-only server preflight before mutation when one exists.
  5. Establish a stable idempotency key for recovery and write a secret-free,
  6. Emit structured stage/code/retryability output. Retain incomplete cleanup
  7. Test at least one negative control proving invalid local input performs zero

What it can do on your machine

Read from SKILL.md and the folder at commit e5d9444. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves these tools, so the agent can use them without asking each time:

    • Read
    • Write
    • Bash
    • WebFetch

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Ships 1 file in scripts/ (Python, from the files we listed), which the agent can run.

    Shell commands in SKILL.md call:

    • node

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

  • Network

    No URLs in SKILL.md.

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

  • Credentials

    Names these keys or tokens, usually read from environment variables:

    • SERVICE_API_KEY

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

  • Compatibility

    ["prismer-sdk","hermes","claude-code","codex","openclaw"]

    From compatibility in the SKILL.md frontmatter.

Context cost

Prismer Skill Creator loads about 2.6k tokens when it runs, and up to ~17k if it reads all its reference files. Until then it costs about 112 tokens; SKILL.md has 983 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~112
When it runs · the whole SKILL.md, loaded when a task matches
~2.6k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~17k

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: notes

The automated check noted patterns worth knowing about, such as sudo or a known installer.

  • NotePre-approves every shell command (allowed-tools: Bash)SKILL.md
    allowed-tools: Read, Write, Bash, WebFetch

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); the scripts in this folder are not scanned.

SKILL.md

The full file from Prismer-AI/PrismerCloud at commit e5d9444, republished under its MIT licence (© Prismer-AI). 983 words, ~2,557 tokens.

Download SKILL.mdSave it as .claude/skills/skill-creator/SKILL.md (or your agent's skills folder). This skill also uses 30 other files; get the full folder from GitHub.
name
skill-creator
description
Create, import, edit, validate, test, or publish a Prismer SS-01 Skill. Use whenever a user asks to turn a repeatable capability or external Skill repository into platform Skills, improve an existing Skill, or make Skills available in Studio, a workspace, or Marketplace. Follow the fixed author-to-readback workflow and use bundled scripts for deterministic repeated work; do not explore raw endpoints or compose ad-hoc CLI chains.
allowed-tools
Read, Write, Bash, WebFetch
compatibility
["prismer-sdk","hermes","claude-code","codex","openclaw"]
scope
common
category
authoring
license
Apache-2.0 AND MIT; see NOTICE.md
metadata.category
authoring
metadata.aliases
skill-builder, skill-authoring
metadata.nativeReplaces
hermes-agent-skill-authoring

Skill Creator

Historical skill-builder and skill-authoring installs resolve to this entry. For low-level legacy formats only, see references/skill-builder/GUIDE.md and references/skill-authoring/GUIDE.md. The current workflow below is authoritative. For Hermes in-repository conventions, read references/hermes-agent-skill-authoring/GUIDE.md only when targeting a verified Hermes checkout. It is a full MIT-licensed reference, not a second entrypoint. The bundled Anthropic authoring resources retain Apache-2.0 in LICENSE.txt.

Choose one route and execute it directly. Do not inspect cloud --help, create probe catalog objects, or call raw Skill endpoints first.

Requested resultRoute
one new or edited SkillSingle bundle
repository/directory containing several Skillsread references/external-library-import.md, then follow it
improve or benchmark a SkillImprove
create a role/persona/job templatestop and use role-builder

Ownership boundary

Optional paid trigger/description evaluations require explicit budget approval and PRISMER_ALLOW_PAID_EVAL=1. Trigger probes use isolated temporary skill directories, Read/Skill only, 0.25 USD per query and a 300-call batch ceiling. Use a current Claude CLI supporting --bare and budget flags; unsupported CLI, timeout, auth failure, or missing terminal result is an error, not a negative trigger. Do not bypass nested-session or authentication guards. Viewer HTML is local; spreadsheet preview additionally uses its declared external SheetJS script and is not guaranteed offline. Static feedback is downloaded explicitly.

Keep each concern in its proper layer:

  • Product resource operations belong in the service/SDK/CLI.
  • A stable, repeated transformation or multi-step workflow belongs in a bundled scripts/ harness.
  • SKILL.md routes intent, declares configuration, invokes the script, and explains the result. It should not make every future Agent rediscover the same command sequence.

Promote a script into the CLI only when several independent clients need the operation as a reusable product primitive. Do not add a CLI verb merely to shorten one Skill workflow.

Single bundle

Create or edit this shape:

text
<skill-slug>/
├── SKILL.md
├── skill.json          # only when samples/security metadata are needed
├── scripts/            # deterministic or repeated work
├── references/         # details read only when needed
└── assets/             # output templates/resources

Minimum SKILL.md:

markdown
---
name: example-skill
description: Describe what it does and concrete situations that should trigger it.
license: MIT
---

# Example Skill

State the shortest reliable workflow, required inputs, output contract, and
failure behavior.

Keep the directory name and frontmatter name equal. Preserve the name when editing an installed Skill. Put detailed variants in references/; keep the triggered body short enough to scan once.

Configuration

Declare reusable Skill configuration in frontmatter and read values only from the script subprocess environment:

yaml
config:
  - key: SERVICE_API_KEY
    type: secret
    required: true
    default: null
    bindable: [global, role, agent]
    description: Credential used by the service client.

Valid types are string, secret, url, and enum. Numeric knobs are strings parsed by the script. Never put a value in SKILL.md, a role, a command argument, a ledger, or model-visible output.

Configuration ownership is strict:

NeedOwnerDelivery
reusable Skill configSkill config:Skill subprocess env
role default/behaviorrole parameters[]structured role params
per-account secretkind:"user", type:"secret"sealed env-only params
role default for a Skillrole skillConfig[slug]resolved Skill env
workspace identity/policyworkspace APIsruntime bootstrap
deployment secretdaemon/deployment envchild process env
Script harness negative controls

When a script writes remotely, it must carry enough harness weight to reject a bad request before the first network call:

  1. Validate required input, types, target, and environment locally.
  2. Reject secret-bearing flags, including --token=value; do not echo the untrusted token or value in an error.
  3. Read credentials and remote-write authority only from trusted environment variables. A flag must not self-authorize a remote write.
  4. Run a read-only server preflight before mutation when one exists.
  5. Establish a stable idempotency key for recovery and write a secret-free, atomic ledger before the first mutation.
  6. Emit structured stage/code/retryability output. Retain incomplete cleanup state instead of claiming success.
  7. Test at least one negative control proving invalid local input performs zero fetches/writes.
Show full SKILL.md (437 more words)Show less

Fixed authoring closure

Run in this order:

bash
cloud skill validate ./<skill-slug> --json
cloud skill test ./<skill-slug> --agent <agentImUserId> --json

validate is local and mandatory. Run test when skill.json declares sampleTasks[]; otherwise it may report a clear skip. Fix failures before any create or publish request.

Choose exactly one destination:

bash
# Private, active, immediately installable in this workspace
cloud skill create ./<skill-slug> --json

# Private draft for Studio human review
cloud skill draft create ./<skill-slug> --json

# Public Marketplace: first create private, then publish returned slug
cloud skill create ./<skill-slug> --json
cloud skill publish <returned-slug> --license <license> --json

Use the server-returned canonical slug. Community Skills may receive a prefix; never reconstruct it from the directory name. Verify private objects through the owner domain, not public search:

bash
cloud skill mine --json
cloud skill show <returned-slug-or-id> --content --json

A 409 is not proof that the intended Skill already exists. Read the owned object and compare it before resuming or choosing a new name. If a documented command is absent, report CLI version skew and use the repository's current Cloud build; do not silently bypass gates with raw HTTP.

External Skill libraries

For more than one source Skill, read references/external-library-import.md, set its small environment contract, and run node scripts/import-library.mjs --json. The harness validates the complete selected tree before writes, rejects mirror slug collisions, resumes through a secret-free ledger, owner-readbacks every canonical slug, and emits the exact requiredSkills array for role-builder. Never replace it with a hand-written shell loop, publish a probe object, or apply a role to the currently executing Agent for verification.

Improve an existing Skill

Default to a small evidence loop, not a large benchmark ceremony:

  1. Preserve the original bundle and name.
  2. Choose 2–3 representative prompts and explicit success checks.
  3. Validate, run, inspect outputs, and revise only observed failure causes.
  4. Re-run the same checks and ask for user review when output quality is subjective.

Only when the user requests a formal benchmark or trigger optimization, use the bundled resources:

  • references/schemas.md for eval/result schemas.
  • scripts/run_eval.py, scripts/aggregate_benchmark.py, and eval-viewer/generate_review.py for comparison/review.
  • scripts/run_loop.py and scripts/improve_description.py for explicit description optimization.
  • agents/grader.md, agents/analyzer.md, and agents/comparator.md only for the corresponding advanced evaluation step.

Do not load all advanced resources for ordinary creation or import.

Built-in baseline boundary

Skill authors define a capability; they do not decide that every Agent must receive it. A community/private Skill cannot self-enrol into the mandatory baseline. A platform Built-in is eligible only through the platform tier registry, and Admin selects/version-controls the active common and orchestration sets in /admin/skills.

Role authors declare only business Skills. Role creation injects the active baseline and exposes requiredSkills, injectedSkills, effectiveSkills, and baselinePolicyVersion. If a proposed baseline Skill requires an unresolved user secret, report it as ineligible; never weaken preflight or embed the value.

Completion report

Report only evidence the user can act on:

  • validation/test result;
  • selected destination and returned canonical slug/version/status/scope;
  • readback result;
  • for batch import, selected/excluded source trees and created/failed ledger counts;
  • any missing configuration by variable name only;
  • whether a role handoff or live Agent mutation was explicitly requested.

© Prismer-AI, 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 30 other files (scripts, references, assets) in sdk/cloud/catalog/skills/skill-creator of Prismer-AI/PrismerCloud.

  • SKILL.md
  • LICENSE.prismer
  • LICENSE.txt
  • NOTICE.md
  • agents/analyzer.md
  • agents/comparator.md
  • agents/grader.md
  • assets/eval_review.html
  • eval-viewer/generate_review.py
  • eval-viewer/viewer.html
  • references/external-library-import.md
  • references/hermes-agent-skill-authoring/GUIDE.md
  • references/hermes-agent-skill-authoring/LICENSE
  • references/schemas.md
  • references/skill-authoring/GUIDE.md
  • … and 16 more

Open the folder on GitHubat commit e5d9444

Compare with similar skills

Prismer Skill Creator 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.

Prismer Skill Creator compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Prismer Skill Creator this skillPrismer-AI/PrismerCloud1.6k—~2.6kAutomated safety check: NotesMIT
Darwin Skill Optimizeralchaincyf/darwin-skill6.2k1 repos~4.7kAutomated safety check: PassMIT
Skill Creatorzhayujie/CowAgent47k—~4.7kAutomated safety check: NotesMIT
Open-Science Skill Creatoraipoch/open-science5.5k—~1.7kAutomated safety check: PassApache-2.0
Skill Quality ReviewerGalaxy-Dawn/claude-scholar5.7k1 repos~3kAutomated safety check: PassMIT
Skill Factorysangrokjung/claude-forge852—~2.7kAutomated safety check: PassMIT

Similar skills

  • Darwin Skill Optimizer

    alchaincyf/darwin-skill

    Scores SKILL.md files on a nine-dimension rubric, then improves them in a keep-or-revert loop with independent judge agents, test prompts, git history and human checkpoints.

    6.2k GitHub starsUsed in 1 repo~4.7k tokens
    Agent WorkflowsAuto-check passed

  • Skill Creator

    zhayujie/CowAgent

    Guides creating, installing and updating agent skills in a workspace: SKILL.md frontmatter, bundled scripts and references, with scripts to scaffold, validate and package.

    47k GitHub stars~4.7k tokensUpdated today
    Agent WorkflowsAuto-check: notes

  • Open-Science Skill Creator

    aipoch/open-science

    Creates, revises, evaluates and publishes skills in the Open-Science app through its native host.skills composer, with optional test prompts and benchmarks.

    5.5k GitHub stars~1.7k tokensUpdated today
    Agent WorkflowsAuto-check passed

  • Skill Quality Reviewer

    Galaxy-Dawn/claude-scholar

    Scores a skill across description, content organization, writing style and structure, then produces letter grades and a prioritized improvement plan.

    5.7k GitHub starsUsed in 1 repo~3k tokens
    Agent WorkflowsAuto-check passed

  • Skill Factory

    sangrokjung/claude-forge

    Analyze session work and automatically convert reusable patterns into Claude Code skills.

    852 GitHub stars~2.7k tokensUpdated 1 mo ago
    Agent WorkflowsAuto-check passed

  • Skill Authoring

    authcrunch/authcrunch.github.io

    Create, port, revise, or audit repo-local skills and their metadata, engineering contracts, and task routes.

    108 GitHub starsUsed in 1 repo~2.5k tokens
    Agent WorkflowsAuto-check passed

More from Prismer-AI/PrismerCloud

All 88 skills in this repo

  • Prismer Google Workspace

    Prismer-AI/PrismerCloud

    Gives an agent account-scoped access to Gmail, Calendar, Drive, Contacts, Docs and Sheets through the gws CLI or a bundled Python client.

    1.6k GitHub starsUsed in 3 repos~4.2k tokens
    Auto-check passed

  • Himalaya Email CLI

    Prismer-AI/PrismerCloud

    Operates a mailbox from the terminal with the external Himalaya CLI over IMAP, SMTP, Notmuch or Sendmail, separate from any built-in email gateway adapter.

    1.6k GitHub starsUsed in 2 repos~2.3k tokens
    Auto-check passed

  • Prismer Image Generation

    Prismer-AI/PrismerCloud

    Generates one image from a text prompt with a bundled Node.js helper and delivers it once as the attachment to the current Prismer reply.

    1.6k GitHub stars~1.4k tokensUpdated 9 days ago
    Auto-check passed

  • Manim Explainer Videos

    Prismer-AI/PrismerCloud

    Produces 3Blue1Brown-style explainer animations with Manim Community Edition for math, algorithms, equations and architecture diagrams, with planning and rendering references.

    1.6k GitHub starsUsed in 2 repos~3.1k tokens
    Auto-check passed

  • Prismer Role Builder

    Prismer-AI/PrismerCloud

    Creates or updates Prismer role templates from a persona, SOP or job description, and turns a role into a working agent that runs its first task through a bundled script.

    1.6k GitHub stars~2.3k tokensUpdated 9 days ago
    Auto-check: notes

  • YouTube Transcript Reformatter

    Prismer-AI/PrismerCloud

    Fetches a YouTube transcript with a helper script and reshapes it into chapters, summaries, X threads, blog posts or timestamped quotes.

    1.6k GitHub starsUsed in 2 repos~905 tokens
    Auto-check passed

Categories

Questions about Prismer Skill Creator

What does Prismer Skill Creator do?

Walks an agent through creating, importing, editing, validating, testing and publishing Prismer Skills with a fixed workflow and bundled scripts. The skill sends each request down one of four routes: a single new or edited skill, a repository or directory holding several skills (which uses a separate import guide), improving or benchmarking an existing skill, or building a role or persona template, which it hands to role-builder. It says to pick a route and act, without first probing cloud --help, creating test catalog objects or calling raw skill endpoints.

When should I use Prismer Skill Creator?

Prismer Skill Creator fits situations like: turning a repeatable capability into a Prismer skill; importing a repository of skills into a workspace or Marketplace; improving or benchmarking an existing Prismer skill; validating a skill bundle before publishing it to Studio.

How do I install Prismer Skill Creator in Claude Code?

Run `npx skills add Prismer-AI/PrismerCloud --skill skill-creator -a claude-code`. Or copy the skill folder (sdk/cloud/catalog/skills/skill-creator in Prismer-AI/PrismerCloud) into .claude/skills/skill-creator in your project. Claude Code loads it when a task matches its description.

How do I install Prismer Skill Creator in Codex?

Run `npx skills add Prismer-AI/PrismerCloud --skill skill-creator -a codex`. Or copy the skill folder (sdk/cloud/catalog/skills/skill-creator in Prismer-AI/PrismerCloud) into .agents/skills/skill-creator in your project. Codex loads it when a task matches its description.

Can I use Prismer Skill Creator 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 Prismer-AI/PrismerCloud --skill skill-creator -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/skill-creator, .gemini/skills/skill-creator, .github/skills/skill-creator and .opencode/skills/skill-creator in your project.

What does Prismer Skill Creator need to run?

Going by SKILL.md and its folder, Prismer Skill Creator needs Python for the scripts in its folder, the command-line tools its instructions call (node) and credentials named SERVICE_API_KEY. Our summary lists: The Prismer SDK or cloud CLI; A current Claude CLI with the --bare flag for optional paid trigger evaluations. Its frontmatter pre-approves these tools: Read, Write, Bash, WebFetch. Compatibility (from SKILL.md): ["prismer-sdk","hermes","claude-code","codex","openclaw"].

Does Prismer Skill Creator access the network?

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.

Is Prismer Skill Creator safe to install?

Our automated static check of SKILL.md found notes only (pre-approves every shell command (allowed-tools: bash)), nothing it rates as a warning. It is not a guarantee. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.

What licence does Prismer Skill Creator use?

Prismer Skill Creator is published under the MIT licence (declared in SKILL.md). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Prismer Skill Creator use?

About 2.6k tokens (SKILL.md is roughly 10k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 14k tokens, read only when the agent opens those files.

What are the alternatives to Prismer Skill Creator?

Skills that share tags, products or a category with Prismer Skill Creator: Darwin Skill Optimizer (alchaincyf/darwin-skill, 6.2k stars), Skill Creator (zhayujie/CowAgent, 47k stars), Open-Science Skill Creator (aipoch/open-science, 5.5k stars) and Skill Quality Reviewer (Galaxy-Dawn/claude-scholar, 5.7k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Prismer Skill Creator?

Prismer-AI (a GitHub organization) maintains it in Prismer-AI/PrismerCloud, which has 1,555 GitHub stars. The repository holds 88 skills in this directory. The repository was last updated on September 30, 2026.

Source: Prismer-AI/PrismerCloud on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.