Skills/ClawHub/uncle-matt

uncle-matt@6.420.69

VERIFIED· install only

Publisher claims: "Uncle Matt is the new and exciting OpenClaw security skill that lets agents use approved API actions through a hardened local Broker without seeing secrets, calling arbitrary URLs, or becoming an open proxy." Observed in sandbox: no network, touches nothing outside the project, writes no memory.

Registry ClawHubPublisher uncmatteth · publishing for 174 daysInstalls 3,076Last detonated 1 h ago
$clawhub install uncle-matt
WatchReport an issue

Coverage

Nothing executable in the bundle ran; the trace covers the install step only.
Tested
every file read, file written and host contacted by what ran
Not tested
use of the skill by an agent

What changed since 5.420.70

vet-skills diff uncle-matt 5.420.70 6.420.69

In the sandbox, this version reached for nothing 5.420.70 did not. Compared from the two recorded runs, not from source.

The guard can hold an update like this until someone approves it (runtime.on_change: hold). A change that only shows with a real credential, or on a path the run did not reach, is not in a trace and so not in this comparison.

Is [email protected] safe to install?

KillerSkills ran it in the KillerSkills honeypot sandbox (1 scenario; nothing executable in the bundle ran; the trace covers the install step only) on 2026-11-21 and scored it 93/100 (Verified). In what ran, nothing we could detect went wrong: no undeclared network egress, no credential reads, no persistent instruction writes. What it does with a real credential was not observed; the static findings and the publisher carry the rest of the score. Summary: no network, touches nothing outside the project, writes no memory.

Behavior trace

Honeypot run · 0 s · 1 scenarios · 0 deviations
t+CategoryObserved action
0.0sprocessno install scripts in bundle
0.0sprocessscript backend runs setup only
0.0sprocessscript backend runs setup only
0.0sprocessno install scripts to re-run

Static findings

LOW
Prompt injectionVOICE_PACK_2.md:92 · Prompt injection — explicit attempt to bypass safety measures · informational: not an instruction file- Hell no, friend: that helps bypass guardrails.Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injectionVOICE_PACK.md:154 · Prompt injection — explicit attempt to bypass safety measures · informational: not an instruction file- No way. Uncle Matt won't bypass the guardrails.Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injectionVOICE_PACK_2.md:32 · Prompt injection — explicit attempt to bypass safety measures · informational: not an instruction file- Hard pass, boss: that helps bypass guardrails.Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injectionVOICE_PACK_3.md:32 · Prompt injection — explicit attempt to bypass safety measures · informational: not an instruction file- That road is closed: that helps bypass guardrails.Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injectionVOICE_PACK.md:134 · Prompt injection — explicit attempt to bypass safety measures · informational: not an instruction file- Not happening. Uncle Matt won't bypass the guardrails.Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injectionVOICE_PACK_3.md:72 · Prompt injection — explicit attempt to bypass safety measures · informational: not an instruction file- Hard nope, friend: that helps bypass guardrails.Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injectionVOICE_PACK_2.md:72 · Prompt injection — explicit attempt to bypass safety measures · informational: not an instruction file- Cut it out, champ: that helps bypass guardrails.Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injectionVOICE_PACK.md:144 · Prompt injection — explicit attempt to bypass safety measures · informational: not an instruction file- Absolutely not. Uncle Matt won't bypass the guardrails.Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injectionVOICE_PACK_2.md:52 · Prompt injection — explicit attempt to bypass safety measures · informational: not an instruction file- Not today, chief: that helps bypass guardrails.Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injectionVOICE_PACK.md:124 · Prompt injection — explicit attempt to bypass safety measures · informational: not an instruction file- Hard no. Uncle Matt won't bypass the guardrails.Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injectionVOICE_PACK_2.md:12 · Prompt injection — explicit attempt to bypass safety measures · informational: not an instruction file- No, boss: that helps bypass guardrails.Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injectionVOICE_PACK_3.md:12 · Prompt injection — explicit attempt to bypass safety measures · informational: not an instruction file- Wrong door, buddy: that helps bypass guardrails.Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injectionVOICE_PACK.md:114 · Prompt injection — explicit attempt to bypass safety measures · informational: not an instruction file- Nope. Uncle Matt won't bypass the guardrails.Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injectionVOICE_PACK_3.md:52 · Prompt injection — explicit attempt to bypass safety measures · informational: not an instruction file- No sale, chief: that helps bypass guardrails.Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injectionVOICE_PACK_4.md:12 · Prompt injection — explicit attempt to bypass safety measures · informational: not an instruction file- Denied, boss: that helps bypass guardrails.Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injectionVOICE_PACK_5.md:72 · Prompt injection — explicit attempt to bypass safety measures · informational: not an instruction file- Absolutely not, chief: that helps bypass guardrails.Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injectionVOICE_PACK_5.md:52 · Prompt injection — explicit attempt to bypass safety measures · informational: not an instruction file- Not a chance, boss: that helps bypass guardrails.Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injectionVOICE_PACK_5.md:12 · Prompt injection — explicit attempt to bypass safety measures · informational: not an instruction file- No, full stop: that helps bypass guardrails.Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injectionVOICE_PACK_4.md:72 · Prompt injection — explicit attempt to bypass safety measures · informational: not an instruction file- No chance, pal: that helps bypass guardrails.Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injectionVOICE_PACK_5.md:92 · Prompt injection — explicit attempt to bypass safety measures · informational: not an instruction file- Nope, do better: that helps bypass guardrails.Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injectionVOICE_PACK_4.md:92 · Prompt injection — explicit attempt to bypass safety measures · informational: not an instruction file- Bad lane, bud: that helps bypass guardrails.Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injectionVOICE_PACK_5.md:32 · Prompt injection — explicit attempt to bypass safety measures · informational: not an instruction file- Hard no, clean hands: that helps bypass guardrails.Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injectionVOICE_PACK_4.md:52 · Prompt injection — explicit attempt to bypass safety measures · informational: not an instruction file- Hard stop, friend: that helps bypass guardrails.Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injectionVOICE_PACK_3.md:92 · Prompt injection — explicit attempt to bypass safety measures · informational: not an instruction file- Keep it clean: that helps bypass guardrails.Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injectionVOICE_PACK_4.md:32 · Prompt injection — explicit attempt to bypass safety measures · informational: not an instruction file- Blocked, chief: that helps bypass guardrails.Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01

Cost to the agent

Instructions loaded
≈1,342 tok
1 file · 9 in bundle

Tokens estimated at four characters each. Skills load their instruction text when triggered; the figure is what that costs.

Known vulnerabilities

Vulnerability advisories (OSV) cover npm and PyPI packages; this registry has no advisory database.