Skill Radar
Use when starting any conversation, before any skill invocation or response. Especially when tempted to invoke a skill for trivial tasks like single-file edits, lookups, or status checks.
Install
$ agentstack add skill-roni-quant-skill-radar-skill-radar ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Links to your public security report.[](https://agentstack.voostack.com/security/report/skill-roni-quant-skill-radar-skill-radar) Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
skill-radar
Lean, signal-driven skill triage. Fires on real triggers, stays silent on noise.
Supersedes upstream using-superpowers. Priority unchanged: user instructions > skills > default system prompt.
Decision Rule
- Match user message against Trigger Taxonomy.
- Match → invoke skill, then act.
- No match → act direct, no skill, no announcement.
- Skill mid-task wrong-fits → say so once, abort skill, continue task.
Trigger Taxonomy
| User signal | Skill | |---|---| | "build", "add feature", "design" | brainstorming → writing-plans | | "fix", "broken", "failing", "bug" | systematic-debugging → TDD | | "refactor large area", "redesign" | writing-plans | | "review", "check code", "PR" | code-review:code-review | | "done", "ship", "merge ready" | verification-before-completion → requesting-code-review | | "write/edit skill" | superpowers:writing-skills | | "create plan", "spec out" | superpowers:writing-plans |
No-Skill Cases (act direct)
- Single-file edit, single-grep, single-command
- Factual questions, lookups, status checks
- Conversational replies, clarifications
- Small refactors (rename, extract in one file)
- Simple bug fixes with obvious cause
Mode Awareness
- Terse / caveman / no-narration modes: skip "Announce: Using X" entirely.
- Heavyweight rigid skills (TDD, debugging): one-line announce OK.
- Subagents: re-state priority hierarchy in subagent prompt — upstream injection gap (obra/superpowers#237) means subagents may not see this.
Composition Chains
- Feature: brainstorming → writing-plans → executing-plans → verification → requesting-code-review.
- Bug: systematic-debugging → TDD → fix → verification.
- Skill author: writing-skills RED → GREEN → REFACTOR.
Stop Conditions
- User says "stop using skills" / "just do it" → respect, no skill that turn.
- Skill body wrong-fits → abort skill, continue task.
Rationalization Table
| Excuse | Reality | |---|---| | "1% chance might apply" | False-positive engine. Match taxonomy or skip. | | "Be safe, invoke anyway" | Burns context. Trust taxonomy. | | "User didn't say it but probably wants TDD" | Taxonomy would match if they did. Don't impose. | | "I'll summarize workflow in description for clarity" | Causes drift — model follows summary, skips body. Triggers-only. |
Example triggers
- "fix this bug" → systematic-debugging
- "rename
fooin user.py" → no skill - "build new export feature" → brainstorming → writing-plans
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: Roni-quant
- Source: Roni-quant/skill-radar
- License: MIT
- Homepage: https://github.com/Roni-quant/skill-radar
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.