Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add yknothing/skills-refiner --skill skill-debuggit clone --depth 1 https://github.com/yknothing/skills-refinerWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/yknothing/skills-refiner/skill-debug)<a href="https://agentmods.dev/skills/yknothing/skills-refiner/skill-debug"><img src="https://agentmods.dev/badge/skills/yknothing/skills-refiner/skill-debug/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/yknothing/skills-refiner/skill-debug"><img src="https://agentmods.dev/badge/skills/yknothing/skills-refiner/skill-debug.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 4 findings, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high Rogue Agent · line 21 Skill modifies its own code, configuration, or behavior at runtime. Self-modification enables an agent to escalate privileges, disable safety constraints, or install persistent backdoors.Fix: Prevent the skill from modifying its own code, SKILL.md, or configuration files. Treat skill files as read-only at runtime.
- medium Agent Snooping · line 67 Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.Fix: Remove all code or instructions that list or read other skills' files or directories. Skills should operate independently; cross-skill access is a privilege escalation.
- medium Agent Snooping · line 113 Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.Fix: Remove all code or instructions that list or read other skills' files or directories. Skills should operate independently; cross-skill access is a privilege escalation.
- medium Agent Snooping · line 119 Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.Fix: Remove all code or instructions that list or read other skills' files or directories. Skills should operate independently; cross-skill access is a privilege escalation.
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00038 | $0.03424 |
| Opus 5.5 | $0.00015 | $0.01370 |
| Sonnet 5.5 | $0.00008 | $0.00685 |
| Haiku 4.5 | $0.00004 | $0.00342 |
Grade A, and why
skill-debug scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 259 lines — stays where its author put it; the contents beside it link to each section on GitHub.
skill-debug
You are a skill observability advisor. Your job is to help users understand what local evidence exists about their skills: likely discovery surfaces, canary activations, and usage patterns. Do not overstate these signals as platform-level proof of loading, obedience, or outcome quality.
Philosophy
- Observe, don't assume. The tools collect facts about skill discovery and activation. You interpret patterns and correlate them with the user's actual experience.
- Respect the topology. Skills are installed to
~/.agents/skills/and symlinked to agent directories. Symlinks pointing to the same source are distribution links, not redundancy. The probe must distinguish symlinks from real duplicates. - No false alarms. A skill with no observed activation may simply not have been needed. "Not observed" is an observation, not a verdict. Cross-reference with the user's actual workflow before recommending removal.
Running tools from an agent session
When you have shell access and the user asks for discovery diagnostics, activation statistics, or a general skills health snapshot:
- Prefer executing read-only scripts yourself (
skills-refiner-doctor.sh,skill-probe.sh,skill-dashboard.sh) rather than instructing the user to run them manually. - Do not run
skill-trace.sh --inject/--inject-dir/--strip/--strip-dirunless the user explicitly asks to modify skills on disk for canary tracing.
One-shot read-only snapshot:
bash ~/.agents/skills/skill-debug/bin/skills-refiner-doctor.sh
bash ~/.agents/skills/skill-debug/bin/skills-refiner-doctor.sh --json
bash ~/.agents/skills/skill-debug/bin/skills-refiner-doctor.sh --lang zh
bash ~/.agents/skills/skill-debug/bin/skills-refiner-doctor.sh --raw
The Problem
Agent skills are "fire and forget" by design. You install them, hope the agent finds them, and have limited ways to verify:
- Is the skill present on a local discovery surface this diagnostic knows how to scan?
- Was the canary command observed during a skill-guided run?
- Did the agent appear to follow the skill's instructions?
- Which installed skill identities have observed canary events vs. no local canary evidence?
What ships with it
15 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- bin/pi-skill-catalog.mjs 6.0 KB runs code
- bin/skill-canary.sh 3.8 KB runs code
- bin/skill-dashboard.sh 16 KB runs code
- bin/skill-probe.sh 21 KB runs code
- bin/skill-trace.sh 17 KB runs code
- bin/skills-refiner-doctor.sh 17 KB runs code
- lib/common.sh 19 KB runs code
- tests/test-dashboard.sh 11 KB runs code
- tests/test-doctor.sh 11 KB runs code
- tests/test-install-layout.sh 43 KB runs code
- tests/test-observability-regressions.sh 5.8 KB runs code
- tests/test-pi-skill-catalog.mjs 7.7 KB runs code
- tests/test-platform-contract.sh 7.5 KB runs code
- tests/test-probe.sh 11 KB runs code
- tests/test-trace.sh 31 KB runs code
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago Changed · +32 lines 559f29dc7568
- 2mo ago Changed · +8 lines b2858bc5df50
- 2mo ago First seen · 219 lines · 38 tokens per session scan A b4606dbafcf2
skill-debug is a skill published in the GitHub repository yknothing/skills-refiner (24 stars, last pushed 1mo ago), licensed MIT. It adds 38 tokens to every session and 3,424 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
skill-debug compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| skill-debugyknothing/skills-refiner | Skill | 24 | 1mo ago | MIT | A | 38 | Claude Code, Codex |
| systematic-debuggingopen-metadata/OpenMetadata | Skill | 16k | today | Apache-2.0 | A | 37 | Claude Code |
| log-error-digestzebbern/claude-code-guide | Skill | 4.7k | today | MIT | A | 71 | Claude Code, Codex |
| cwv-optimizeradobe/skills | Skill | 326 | today | Apache-2.0 | A | 112 | Claude Code |
| ci-fixersipyourdrink-ltd/bernstein | Skill | 1.2k | today | Apache-2.0 | A | 15 | Claude Code, Codex |
| diagnosing-missing-recordingsPostHog/posthog | Skill | 41k | today | none | A | 76 | Claude Code, Codex |
| triaging-web-analytics-supportPostHog/posthog | Skill | 41k | today | none | A | 151 | Claude Code, Codex |