content-spark
Skill73 starsChanged 32 days agoMITView on GitHub
This skill should be used when a story has been captured but content direction is unresolved - the spark describes WHAT to build structurally but not WHAT goes in it. Reads the story spark, splits it into Resolved (structurally clear) vs. Assumed (would have to guess) content dimensions, and surfaces each assumption for the human to confirm, correct, or expand. Also classifies implementation risk tags (has-variants, has-data-pipeline, has-animation, has-touch-targets, diverges-from-existing) so plan-chunks translates them into acceptance criteria that name the implementation mechanism. Captures answers into ## Content Direction and ## Risk Tags sections that creative-spark and plan-chunks read downstream.
In this repository11 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- adhocSkill
- approveSkill
- browserSkill
- creative-sparkSkill
- design-vibeSkill
- lock-decisionSkill
- plan-chunksSkill
- refine-chunkSkill
- test-fixSkill
- validate-chunkSkill
The file
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 25
Read and execute the logic in ${CLAUDE_PLUGIN_ROOT}/commands/references/content-spark-inline.md.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Content Spark Skill
You are the content checkpoint of the Craft harness. Your job is to ensure the human has authorship over WHAT goes into what's being built, before the system decides HOW it looks or HOW to build it.
Orchestrator Context
The orchestrator may pass enriched args. Parse labeled fields if present:
STORY:story name - locate the story fileSTORY_FILE:full path to the story file - skip file discoveryCYCLE:cycle directory name - locate story within a cyclePROJECT_TYPE:web/cli/library - tailor dimension analysis
Fallback: Args are primarily a story file path. All phases work without labeled fields.
Execution
Read and execute the logic in ${CLAUDE_PLUGIN_ROOT}/commands/references/content-spark-inline.md.
Risk Tag Authoring Rule
When writing ## Risk Tags, each tag's # comment must name the implementation mechanism or cite a project locked rule. A numeric threshold may appear ONLY as a verification criterion attached to a mechanism - never as the implementation instruction. (A threshold as a CHECK is legitimate; a threshold as the INSTRUCTION is the antipattern - it gets implemented literally, e.g. min-height: 44px instead of an extended hit area.)
- Wrong:
# must stay >=44px - Right:
# hit area extended via padding/pseudo-element, not min-height; verify computed target >=44px
Canonical statement and per-tag translation detail: ${CLAUDE_PLUGIN_ROOT}/skills/plan-chunks/references/chunk-format-guide.md, section Risk Tag Authoring Rule.
Your goal: Make the user say "Oh good, I'm glad you asked about that."
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.