promptherder
Syncs AI agent configurations and workflows from shared repositories to local project targets.
Install
mkdir -p .claude/skills/promptherder && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/11791" && unzip -o skill.zip -d .claude/skills/promptherder && rm skill.zipInstalls to .claude/skills/promptherder
Activation
This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.
Reference for the promptherder CLI — syncs AI agent rules, skills, and workflows from herd repos to agent targets. Use when installing, updating, or troubleshooting prompt distribution.Key capabilities
- →Install shared AI agent skills from GitHub herds
- →Sync AI agent rules and workflows to local targets
- →Merge configurations from herds and local files
- →Track file ownership in a manifest
- →Preview synchronization changes without writing
- →Sync specific agent targets
How it works
The skill pulls shared skills and rules from GitHub herds and merges them with local configuration files. It then distributes the combined configuration to various AI agent targets, tracking file ownership in a manifest.
Inputs & outputs
When to use promptherder
- →Install shared AI agent skills from a herd
- →Sync agent rules to local project targets
- →Troubleshoot prompt distribution issues
- →Update AI workflows across repos
About this skill
Promptherder 1.x
Select targets explicitly: promptherder install codex claude, or interactive install with nothing preselected. install none saves no targets. Use target list, target add, and target remove; run bare promptherder to apply the selection. No host is a default.
pull <alias-or-url> downloads a herd without syncing. Sources remain in .promptherder/herds/; local additions live in .promptherder/agent/. Differing local overrides require source paths in settings overrides. .promptherder/hard-rules.md joins the baseline on sync. Edit sources, never generated host files.
Use plan --json to inspect content, origins, ownership, deletions, and diagnostics; check to validate; doctor for additional personal discovery checks; explain <id> for source tracing. The documented profile is not proof of runtime adherence.
Codex receives AGENTS.md and .agents/skills/; Claude receives a CLAUDE.md import bridge, .claude/rules/, and .claude/skills/. Other repository targets are Copilot, Cursor, Windsurf, Cline, and Antigravity. Gemini CLI is removed. Native workflow IDs are workflow-<name> (plus any configured prefix): $workflow-plan in Codex, /workflow-plan in Claude. Skill directories preserve resources and executable scripts; herd.json.skills can declare helper dependencies and host overlays.
Review legacy migration with plan --adopt --json before syncing with --adopt. User edits to tracked output are conflicts, not overwrite permission. After an interrupted apply, recover checks for subsequent edits before rollback.
Lockfiles record immutable revisions and content digests. Review changed herds with plan --update-lock, then apply --update-lock; pull --locked reproduces a recorded revision. Direct target sync rebuilds the installed consumer union without changing saved preferences.
--scope user explicitly selects personal installation for Codex/Claude in default layouts, with sources and settings under ~/.promptherder/. Repository scope remains the default. Use the same scope for setup, pull, sync, and recovery. Read the installed version's help before relying on these commands; @latest can lag an unreleased feature branch.
When not to use it
- →When the task is not about managing AI agent configuration
- →When the user is not installing, updating, or troubleshooting prompt distribution
- →When two herds provide the same file path, causing a conflict error
Limitations
- →Two herds providing the same file path will result in a conflict error
- →Files listed in the manifest's 'generated' array are never overwritten
- →Pulling a herd does not automatically sync; a separate sync command is required
How it compares
This skill centralizes and automates the distribution of AI agent configurations from multiple sources to various targets, which is more efficient than manually managing individual agent setups.
Compared to similar skills
promptherder side by side with the closest alternatives in the catalog.
| Skill | Installs | Updated | Safety | Difficulty |
|---|---|---|---|---|
| promptherder (this skill) | 0 | 5mo | Review | Intermediate |
| prompt-optimizer | 43 | 7mo | No flags | Beginner |
| context-compression | 13 | 4mo | Review | Advanced |
| learner | 2 | 5mo | No flags | Advanced |
Try saying
Example prompts that trigger this skill in your AI assistant.
More by shermanhuman
View all by shermanhuman →You might also like
prompt-optimizer
solatis
Optimize system prompts for Claude Code agents using proven prompt engineering patterns. Use when users request prompt improvement, optimization, or refinement for agent workflows, tool instructions, or system behaviors.
context-compression
muratcankoylan
This skill should be used when the user asks to "compress context", "summarize conversation history", "implement compaction", "reduce token usage", or mentions context compression, structured summarization, tokens-per-task optimization, or long-running agent sessions exceeding context limits.
learner
Yeachan-Heo
Extract a learned skill from the current conversation
sub-agents
parcadei
Create and configure Claude Code sub-agents with custom prompts, tools, and models
skillforge
tripleyak
Intelligent skill router and creator. Analyzes ANY input to recommend existing skills, improve them, or create new ones. Uses deep iterative analysis with 11 thinking models, regression questioning, evolution lens, and multi-agent synthesis panel. Phase 0 triage ensures you never duplicate existing functionality.
reflect
llama-farm
Analyze the current session and propose improvements to skills. **Proactively invoke this skill** when you notice user corrections after skill usage, or at the end of skill-heavy sessions. Also use when user says "reflect", "improve skill", or "learn from this".