Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/softspark/ai-toolkitnpx agentmods add skills/softspark/ai-toolkit/skill-auditWrote 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/softspark/ai-toolkit/skill-audit)<a href="https://agentmods.dev/skills/softspark/ai-toolkit/skill-audit"><img src="https://agentmods.dev/badge/skills/softspark/ai-toolkit/skill-audit/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/softspark/ai-toolkit/skill-audit"><img src="https://agentmods.dev/badge/skills/softspark/ai-toolkit/skill-audit.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 9 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 Tool Misuse · line 61 Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).Fix: Validate all tool parameters against an allowlist. Reject dangerous parameter values (shell=True, --force, -rf /) and use safe defaults.
- high Tool Misuse · line 61 Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).Fix: Validate all tool parameters against an allowlist. Reject dangerous parameter values (shell=True, --force, -rf /) and use safe defaults.
- high Tool Misuse · line 61 Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).Fix: Validate all tool parameters against an allowlist. Reject dangerous parameter values (shell=True, --force, -rf /) and use safe defaults.
- high Tool Misuse · line 61 Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).Fix: Validate all tool parameters against an allowlist. Reject dangerous parameter values (shell=True, --force, -rf /) and use safe defaults.
- high Tool Misuse · line 63 Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).Fix: Validate all tool parameters against an allowlist. Reject dangerous parameter values (shell=True, --force, -rf /) and use safe defaults.
- medium Excessive Agency · line 39 Skill grants unrestricted tool access without appropriate constraints. An agent with unfettered tool access can perform arbitrary actions including file modification, network requests, and code execution.Fix: Restrict tool access to only the tools required for the skill's stated purpose. Use an explicit allowlist rather than granting blanket access.
- medium Privilege Escalation · line 63 Commands invoke sudo or root privileges. Verify this elevated access is necessary and justified.Fix: Avoid sudo/root unless strictly required. Prefer least-privilege patterns. If elevation is needed, document the justification and scope.
- medium Excessive Agency · line 104 Skill grants unrestricted tool access without appropriate constraints. An agent with unfettered tool access can perform arbitrary actions including file modification, network requests, and code execution.Fix: Restrict tool access to only the tools required for the skill's stated purpose. Use an explicit allowlist rather than granting blanket access.
- medium Excessive Agency · line 147 Skill grants unrestricted tool access without appropriate constraints. An agent with unfettered tool access can perform arbitrary actions including file modification, network requests, and code execution.Fix: Restrict tool access to only the tools required for the skill's stated purpose. Use an explicit allowlist rather than granting blanket access.
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 | $0.00036 | $0.01801 |
| Opus 5.5 | $0.00014 | $0.00720 |
| Sonnet 5.5 | $0.00007 | $0.00360 |
| Haiku 4.5 | $0.00004 | $0.00180 |
Grade B, and why
skill-audit scanned grade B with 4 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.
Asks for rootlowPrivilege escalation
A mod that escalates privileges can change anything on the machine, not only the project.
| `chmod 777` | WARN | Overly permissive | Downgraded: this mod is about security review, or the phrase is quoted, so it is likely naming the pattern rather than instructing it.
Recursive force deletemediumDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
| `rm -rf /` or `rm -rf ~` | HIGH | Destructive command | Downgraded: this mod is about security review, or the phrase is quoted, so it is likely naming the pattern rather than instructing it.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
| `curl.*\| bash` | HIGH | Remote code execution | Runs shell commandslowCapability
Expected in a hook, worth knowing in a rule or an instructions file.
| `os.system(` | HIGH | Shell injection risk | How it starts
The opening of the file, as written. The whole thing — 165 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/skill-audit - Skill Security Scanner
$ARGUMENTS
Scan skill and agent definitions for security risks before installation or after changes.
Usage
/skill-audit # Audit all skills
/skill-audit debug # Audit specific skill
/skill-audit --all --fix # Audit all + auto-fix safe issues
What This Command Does
- Scan SKILL.md frontmatter for permission issues
- Scan scripts/ for dangerous code patterns
- Scan reference/ for hardcoded secrets
- Report findings with severity levels
- Auto-fix safe issues when
--fixis passed
Security Checks
Frontmatter Checks
| Check | Severity | Description |
|---|---|---|
| Overly permissive tools | WARN | allowed-tools includes Bash + Write + Edit without justification |
| Missing allowed-tools | WARN | No tool restriction = full access |
| Knowledge skill with Bash | HIGH | user-invocable: false skills should not need Bash |
| Missing effort field | INFO | Best practice to declare effort |
Script Checks (Python)
| Pattern | Severity | Description |
|---|---|---|
eval( / exec( |
HIGH | Arbitrary code execution |
os.system( |
HIGH | Shell injection risk |
subprocess.*shell=True |
HIGH | Shell injection risk |
__import__ |
WARN | Dynamic imports |
pickle.loads |
HIGH | Deserialization attack |
open(.*'w') without path validation |
WARN | Arbitrary file write |
Script Checks (Bash)
| Pattern | Severity | Description |
|---|---|---|
curl.*| bash |
HIGH | Remote code execution |
curl.*| sh |
HIGH | Remote code execution |
wget.*| bash |
HIGH | Remote code execution |
rm -rf / or rm -rf ~ |
HIGH | Destructive command |
Unquoted $variables in commands |
WARN | Word splitting / injection |
chmod 777 |
WARN | Overly permissive |
Secret Detection
| Pattern | Severity | Description |
|---|---|---|
AKIA[0-9A-Z]{16} |
HIGH | AWS access key |
sk-[a-zA-Z0-9]{20,} |
HIGH | API key pattern |
password\s*=\s*['"][^'"]+ |
WARN | Hardcoded password |
token\s*=\s*['"][^'"]+ |
WARN | Hardcoded token |
-----BEGIN.*PRIVATE KEY |
HIGH | Private key |
ghp_[a-zA-Z0-9]{36} |
HIGH | GitHub PAT |
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 First seen · 165 lines · 36 tokens per session scan B 0b4e54a67749
skill-audit is a skill published in the GitHub repository softspark/ai-toolkit (180 stars, last pushed yesterday), licensed Apache-2.0. It adds 36 tokens to every session and 1,801 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it B with 4 findings (asks for root, recursive force delete, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other skills, from other repositories
cline-fix-volatile-msg
Ladder-aware Cline Anthropic caching — verify the rolling read/write ladder on the wire, then add the tools breakpoint and tune TTL. Updated for the 2026-08 AI-SDK monorepo.
continue-gemini-explicit
Continue's Gemini provider doesn't use the cachedContents API at all. Add explicit caching for sessions over the minimum token threshold.
roo-fix-volatile-msg
Ladder-aware Roo Code Anthropic caching — verify the rolling read/write ladder on the wire, then close the real gaps (Vertex 4-block budget, MiniMax path).
aider-1h-ttl
Aider uses 5min TTL by default and works around long pauses with keepalive pings. Wire up the 1h TTL beta instead.
cline-openai-cache-key
Cline OpenAI native provider sends no promptcachekey. Add a stable per-task key so cachedtokens stops being zero.
continue-enable-defaults
Continue's prompt caching is opt-in via config and off by default. Flip the default to systemAndTools.