agentsclimarketplace

Skill generator

Skill Razaib-khan/ForgeWeave/src/forgeweave/templates/claude/.claude/skills/skill-generator

Behavioral execution framework for AI agents — define deterministic, portable skills & agents across OpenCode, Claude Code, Gemini CLI, and Qwen Code.

Install
npx -y skills add Razaib-khan/ForgeWeave --skill skill-generator

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Meta-skill that creates new SKILL.md files from research outputs and patterns discovered in execution

SKILL.md

2.8 KB, as published. Nobody here has run it

Skill Generator

Purpose

Analyze execution patterns, research findings, or user workflows and distill them into reusable ForgeWeave skills. Identifies repeatable jobs, extracts the workflow steps, and produces a complete SKILL.md bundle with scripts, references, and evals.

When to Use

  • A workflow has been repeated 2+ times manually
  • Research findings reveal a repeatable process that should be codified
  • An ad-hoc process in a session should become a reusable skill
  • The user says "we should make this a skill"

When Not to Use

  • The workflow is a one-time task — skip
  • The workflow requires no agent judgment (use a regular script)
  • Creating a skill for an existing skill (use the skill-authoring skill instead)

Inputs

InputTypeRequiredDescription
sourcestringYesDescription of the repeatable job or path to research output
namestringYesDesired skill name (kebab-case)
workflow_stepsarrayNoExisting workflow notes if available

Expected Outputs

OutputDescription
SKILL.mdComplete skill bundle following the 10-section format
scripts/Deterministic steps extracted from the workflow
evals/Validation tests for the skill
references/Supporting documentation

Exact Workflow Steps

  1. Analyze the source workflow or research to identify the repeatable job
  2. Scope: one job per skill, max 3 use cases
  3. Write SKILL.md with all 10 sections in order
  4. Extract deterministic steps into standalone scripts
  5. Create eval tests for happy path and failure modes
  6. Validate: bundle structure, frontmatter, section order, scripts exist

Required Checks

  • Skill covers exactly one repeatable job
  • SKILL.md has all 10 sections in order
  • All referenced scripts exist
  • At least one eval covers the happy path

Failure Modes

Failure ConditionResponse
Source workflow is too broadNarrow to one repeatable job
No deterministic steps foundSkill is fine with SKILL.md only
Validation failsFix issues and regenerate

Examples

Example: Generate a skill from research output

  1. Research on "Next.js 16 caching APIs" reveals a repeatable pattern
  2. Name: nextjs-caching
  3. SKILL.md captures: purpose, workflow steps for analyzing caching configs, inputs
  4. evals/ test the skill against known Next.js 16 patterns

References

ReferencePath
Skill Authoring skill../skill-authoring/SKILL.md
SKILL_SPEC.md./SKILL_SPEC.md

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.