Skillcaller — Development skill for Claude Code
Trigger-reliability eval harness for Agent Skills: proves your skills fire when they should and stay silent when they shouldn't.
How to install Skillcaller
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open iVamsi/skillcaller and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Skillcaller does
Trigger-reliability eval harness for Agent Skills: proves your skills fire when they should and stay silent when they shouldn't.
Alternatives in Development
- Aod.Constitution — Create or update the project constitution from interactive or provided principle inputs, ensuring all dependen 89 ★
- Botbus Protocol — The BotBus protocol: how a menu bar app on your computer and the companion apps on your phone and watch stay i 81 ★
- Fire Cost — Track context usage, estimate session costs, and trigger strategic compaction 77 ★
README
skillcaller
[](https://github.com/iVamsi/skillcaller/actions/workflows/ci.yml) [](https://scorecard.dev/viewer/?uri=github.com/iVamsi/skillcaller) [](LICENSE) [](package.json)
Trigger-reliability evals for [Agent Skills](https://agentskills.io). Proves your skills fire when they should, stay silent when they shouldn't, and don't steal each other's prompts.
npx skillcaller run
PASS writing-compose-previews triggers 100% false triggers 0%
FAIL writing-compose-ui triggers 80% false triggers 0%
- trigger rate 80% is below the gate of 90%
Collisions (another skill answered these prompts):
writing-compose-ui -> answered by writing-compose-previews 20% of the time
1 of 2 skill(s) failed. cost $0.31
Quick start
1. Installation
Run directly with `npx` (no installation required), or add it to your project:
# Run directly:
npx skillcaller run
# Or install as a development dependency:
npm install -D skillcaller
2. Where skills live (Auto-discovery)
`skillcaller` zero-configures by auto-detecting skills in any of these standard directories:
./skills(general project skills).agents/skills(standard Agent Skills directory for Cursor & Codex).claude/skills(Claude Code directory).cursor/skills(Cursor directory)
You can also pass any custom directory path explicitly:
npx skillcaller run path/to/my-skills
3. Step-by-step setup
Step A: Scaffold an eval corpus for a skill
npx skillcaller init skills/my-skill
This creates `skills/my-skill/evals/triggers.yaml` ne
Related Skills
Skill Anthropic Grade Optimizer
Audits Claude-directing artifacts (CLAUDE.md, SKILL.md, subagent, hook, MCP, prompts, api_config) against 189
Auto Trigger
After any recon/probe step, route observed signals to the right tools and fire the auto-plan (Balanced policy)
Auto Explore
Manually trigger auto-explore on a specific file (Desktop fallback when PostToolUse hook does not fire per RES
Unchatty
Make Claude Code write plainly. A CLAUDE.md block plus the A/B harness that proves it works.
Hermesbench
Reliability-first benchmark and evaluation harness for Hermes Agent runtime configurations
Agentpack Skill Review
Review an existing agent skill and generate a balanced trigger/non-trigger eval set with a reproducible iterat
Related Agents
Arch Evaluator
Runs the eval harness against a newly generated plugin — skill trigger accuracy, agent delegation correctness,
Eval Code
General-purpose code reviewer. Synthesizes every eval- lens (correctness, types, async, security, tenancy, dat
Harness Code Reviewer
Code reviewer — two-stage review against a pinned SHA: spec compliance first, then code quality, hunting fail-