SKILLEMALL.ai
GitHub ↗ RU Sign in

FF chain

Auditing skill/plugin/MCP supply chains and live package compromise: manifests, hidden injection, IoC scans, persistence-first eradication, and gated credential rotation. Not for app SAST (Sentinel).

Not recommendedcritical or high security findings · low grade F
simota/agent-skills Agent Skills author: simota 12 files body ≈ 5 859 tokens Open the sourcegithub.com↗ analyzed 80 min ago

Auditing skill/plugin/MCP supply chains and live package compromise: manifests, hidden injection, IoC scans, persistence-first eradication, and gated…

As a process F 45/100 · Will not run — References files that are not bundled: _common/SECURITY.md, _common/OPERATIONAL.md

AnalyzerAI and agentsSecuritytype and topics are labelled automatically from the skill text
Runs in: Claude / Claude Code ChatGPT / Codex Gemini CLI Cursor GitHub Copilot / VS Code Qwen Code Kimi Code CLI Z.ai (GLM) DeepSeek (Deep Code) Mistral Vibe LangChain (deepagents) OpenClaw Goose OpenCode
JSON
Technical rating
F
26/100
safety, quality, tests
Safety 60%
0
Quality 40%
64
Run on models
none yet
Process rating
F
45/100
Will not run
References files that are not bundled: _common/SECURITY.md, _common/OPERATIONAL.md
Tools and files w 18
0
Inputs and preconditions w 11
0
Failures and branches w 10
0
the three weakest of ten parameters · all ten

What is at stake

The skill contains fragments that, in the wrong hands, cost money or data. Below: what the installer risks and what the author should do.

Dangerous commands
If you install

The skill contains commands that delete files, rewrite disks or execute code fetched from the network. The agent may run them without asking if it believes the instructions require it.

For the author

Replace destructive commands with safe equivalents that ask for confirmation, scope them to one folder, and stop piping curl into a shell: pin a version and a checksum.

Obfuscation
If you install

The files contain invisible characters, encoded commands or comments hidden from readers but visible to the model. What you read differs from what the agent sees.

For the author

Remove invisible characters (they usually sneak in through copy-paste) and encoded strings: no catalog will pass them. Instructions for the model must be readable by a human too.

Risky intent medium severity

Below is the worst case for this category. The finding here is medium: the guard saw a sign, not a proof.

If you install

The purpose itself is risky: wallets, browser password stores, offensive security. Even an honest implementation gives the agent access to things that cost money.

For the author

Explain in the description why the access is needed and how it is limited; add tests that show refusals on dangerous requests.

How to improve

  1. Address the high-severity findings: each costs 18 safety points. If one is a false positive, add the rule id to guard.allow in spec.yaml.
  2. Say in the description WHEN to use the skill ("use when…", example requests): that is the agent's main cue.
  3. The SKILL.md body is over 5,000 tokens: move reference detail into references/ and load it when needed.
  4. The text references files that are not there: add them or drop the references.
For the model run — optional
  • Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
  • A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.

Guard findings · 35

  • high Dangerous commands cmd-destructive-fs reference/supply-chain-malware-ioc-database.md:143
    Destructive filesystem command (wipes root/home/drive) (documentation of a security skill)
    rm -rf ~/
    security skill
  • high Obfuscation uni-bidi-override reference/unicode-tag-scan.md:144
    Bidirectional override characters (Trojan Source) (2 occurrences)
    ␀This Arabic line uses RLE intentionally.␀
Medium and low: 33
  • medium Dangerous commands cmd-encoded-exec reference/bundled-artifact-review.md:29
    Executes a base64/encoded payload (documentation table row; documentation of a security skill)
    | `base64 -d \| sh` / `base64 \| bash` | `P0` | hidden payload |
    tablesecurity skill
  • medium Dangerous commands cmd-persistence reference/supply-chain-malware-eradication.md:61
    Persistence mechanism (cron / launchd / scheduled task / autorun registry) (documentation of a security skill)
    launchctl unload ~/Library/LaunchAgents/com.user.gh-token-monitor.plist
    security skill
  • medium Dangerous commands cmd-persistence reference/supply-chain-malware-eradication.md:62
    Persistence mechanism (cron / launchd / scheduled task / autorun registry) (documentation of a security skill)
    launchctl bootout gui/$(id -u) ~/Library/LaunchAgents/com.user.gh-token-monitor.plist 2>/dev/null
    security skill
  • medium Dangerous commands cmd-persistence reference/supply-chain-malware-eradication.md:126
    Persistence mechanism (cron / launchd / scheduled task / autorun registry) (documentation of a security skill)
    ~/Library/LaunchAgents/com.user.gh-token-monitor.plist   # macOS
    security skill
  • medium Risky intent intent-wallet-secrets reference/supply-chain-malware-eradication.md:276
    Handles crypto-wallet secrets (seed / mnemonic / private key) — a classic stealer target
    | 20 | Treat the compromised seed phrase as burned — never reuse |
  • medium Risky intent intent-wallet-secrets reference/supply-chain-malware-eradication.md:324
    Handles crypto-wallet secrets (seed / mnemonic / private key) — a classic stealer target
    | Reuse wallet seed after "cleaning" the machine | Seed phrase is burned the moment it left the host | New wallet, always |
  • medium Dangerous commands cmd-persistence reference/supply-chain-malware-ioc-database.md:116
    Persistence mechanism (cron / launchd / scheduled task / autorun registry) (documentation of a security skill)
    ~/Library/LaunchAgents/com.user.gh-token-monitor.plist
    security skill
  • medium Dangerous commands cmd-persistence reference/supply-chain-malware-scan-procedures.md:60
    Persistence mechanism (cron / launchd / scheduled task / autorun registry) (documentation of a security skill)
    ls -la ~/Library/LaunchAgents/ 2>/dev/null
    security skill
  • medium Dangerous commands cmd-persistence reference/supply-chain-malware-scan-procedures.md:61
    Persistence mechanism (cron / launchd / scheduled task / autorun registry) (documentation of a security skill)
    ls -la /Library/LaunchAgents/  2>/dev/null
    security skill
  • medium Dangerous commands cmd-persistence reference/supply-chain-malware-scan-procedures.md:68
    Persistence mechanism (cron / launchd / scheduled task / autorun registry) (documentation of a security skill)
    plutil -p ~/Library/LaunchAgents/<file>.plist
    security skill
  • medium Dangerous commands cmd-encoded-exec SKILL.md:191
    Executes a base64/encoded payload (documentation table row; documentation of a security skill)
    | `base64 -d \| sh` / `base64 \| bash` | hidden payload execution |
    tablesecurity skill
  • low Dangerous commands cmd-pipe-to-shell reference/bundled-artifact-review.md:26
    Downloads and executes remote code from an unrecognised host (pipe to shell) (documentation table row; documentation of a security skill)
    | `wget -qO- ... \| sh` | `P0` | unpinned remote execution |
    tablesecurity skill
  • low Risky intent intent-offensive-security reference/bundled-artifact-review.md:31
    Offensive-security / dual-use content (legitimate for authorised testing; review intended use)
    | `nc -e`, `ncat -e`, `bash -i >& /dev/tcp/` | `P0` | reverse shell |
  • low Dangerous commands cmd-destructive-fs-scoped reference/bundled-artifact-review.md:33
    Destructive command on a partition / current directory (normal in admin docs; verify the target) (documentation table row; documentation of a security skill)
    | `rm -rf $HOME`, `rm -rf /`, `dd if=/dev/zero of=/dev/sda` | `P0` | destructive |
    tablesecurity skill
  • low Exfiltration exfil-read-secret-files reference/bundled-artifact-review.md:34
    Reads credential / secret files (documentation table row; documentation of a security skill)
    | `cat ~/.ssh/id_*`, `cat ~/.aws/credentials`, `cat ~/.netrc` | `P0` | credential exfil |
    tablesecurity skill
  • low Exfiltration exfil-read-secret-files reference/bundled-artifact-review.md:35
    Reads credential / secret files (documentation table row; documentation of a security skill)
    | `cat ~/.config/gh/hosts.yml`, `gh auth token` | `P0` | GitHub PAT exfil |
    tablesecurity skill
  • low Dangerous commands cmd-persistence reference/supply-chain-malware-ioc-database.md:244
    Persistence mechanism (cron / launchd / scheduled task / autorun registry) (documentation table row; documentation of a security skill)
    | macOS LaunchAgent | `~/Library/LaunchAgents/*.plist` with `ProgramArguments` referencing unfamiliar interpreter (`bun`, `deno`, hidden node path) |
    tablesecurity skill
  • low Dangerous commands cmd-cron-mention reference/supply-chain-malware-ioc-database.md:247
    Mentions editing / listing crontab (documentation table row; documentation of a security skill)
    | Linux cron | `crontab -l` entries referencing `~/.config/`, `/tmp/`, or `~/.local/bin/` |
    tablesecurity skill
  • low Dangerous commands cmd-persistence reference/supply-chain-malware-ioc-database.md:249
    Persistence mechanism (cron / launchd / scheduled task / autorun registry) (documentation table row; documentation of a security skill)
    | Windows Run keys | `HKCU\Software\Microsoft\Windows\CurrentVersion\Run` entries added recently |
    tablesecurity skill
  • low Dangerous commands cmd-persistence reference/supply-chain-malware-ioc-database.md:309
    Persistence mechanism (cron / launchd / scheduled task / autorun registry) (documentation table row; documentation of a security skill)
    | `~/Library/LaunchAgents/com.user.gh-token-monitor.plist` | Mini Shai-Hulud 2nd persistence | `launchctl unload` **before** any token revoke |
    tablesecurity skill
  • low Dangerous commands cmd-cron-mention reference/supply-chain-malware-scan-procedures.md:140
    Mentions editing / listing crontab (documentation of a security skill)
    crontab -l 2>/dev/null
    security skill
  • low Dangerous commands cmd-pipe-to-shell SKILL.md:192
    Downloads and executes remote code from an unrecognised host (pipe to shell) (documentation table row; documentation of a security skill)
    | `curl ... \| bash`, `wget ... \| sh` | unpinned remote code execution |
    tablesecurity skill
  • low Risky intent intent-offensive-security SKILL.md:196
    Offensive-security / dual-use content (legitimate for authorised testing; review intended use)
    | `nc -e`, `bash -i >& /dev/tcp` | reverse shell |

A further 10 matches are quotations in this security skill's documentation and are not counted as findings.

Files scanned: 12. Evidence is masked. Grey chips explain why severity was lowered.

Against the Agent Skills spec

  • warning description-no-when description does not say WHEN to use the skill (no "use when")
  • warning body-long SKILL.md body ≈ 5859 tokens (recommended < 5000); move details to references/
  • warning missing-ref reference to a missing file: _common/SECURITY.md
  • warning missing-ref reference to a missing file: _common/OPERATIONAL.md

Process rating: all ten parameters 45/100

Will not run. References files that are not bundled: _common/SECURITY.md, _common/OPERATIONAL.md
  • 0Tools and files. 2 referenced file(s) missing: _common/SECURITY.md, _common/OPERATIONAL.md
  • 0Inputs and preconditions. Does not say what the process needs to start
  • 0Failures and branches. Linear process with no failure handling
  • 30Running it twice. 8 mutating operations with no state check
  • 50When it triggers. No condition that starts the skill
  • 60Result and completion. Output format stated, no completion criterion
  • 70Execution cost. Instruction body is 5859 tokens
  • 100Steps. 104 steps
  • 100Consistency. Name and required fields are in place
  • 100Progress reporting. Reports progress
  • medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
  • low 15 top-level sections: this looks like several domains in one skill

Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.

Quality signals

  • +5Description has no quoted example phrases that should trigger the skill
  • +1No license
  • +2Single-language instructions
  • +4Description says when NOT to use the skill
  • +3Description length 199: enough signal without eating the budget
  • +4Structure: 21 headings
  • +3Step-by-step instructions: 104 items
  • +3Output format is stated explicitly
  • +4Has examples (1 code blocks)
  • +4Reference files are cited in the instructions (11 of 11)

Quality base 70; lint remarks subtract, signals add up to 100. Result: 64.