agent-assets— agent skill

Cordfuse agent skills for the Agent Skills Directory (agentskills.io)

by cordfuse·Agent Skill·github.com/cordfuse/agent-assets

Is agent-assets safe to install?

SaferSkills independently audited agent-assets (Agent Skill) and scored it 79/100 (yellow). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 5 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.

Score
79/100
●●●●●●●●○○
↑ +0 since first scan (79 → 79)Re-scan~30s
Latest scan
ScannedJun 28, 2026 · 26d ago
Scans run1 over 90 days
Detectors55 checks · 5 categories
Findings5 warnings · 0 high
EngineSaferSkills 2b638c6
View methodology →
SaferSkills installs
This week0
This month0
All time0
CategoryWeightCategory scoreContribution
Securityprompt, exec, net, exfil, eval
35%
40
14.0 pts
Supply chainhash, typosquat, maintainer, lockfile
20%
100
20.0 pts
Maintenancestaleness, pinning, CI
15%
100
15.0 pts
TransparencySKILL.md, perms, README
15%
100
15.0 pts
Communityinstalls, verify, response
15%
100
15.0 pts

Findings & checks · 5 flagged

Securityscore 40 · 5 findings
MEDIUMInstruction telling the agent not to ask for approvalSS-SKILL-INJECT-DONT-ASK-01 · Prompt injection · actors/guitar-tone-advisor.md×2
MEDIUMit fires on intent; the real damage depends on the host agent's own approval-gating.
Why it matters

The text {match} tells the agent to skip the normal "ask the user first" gate. Used adversarially it removes the human-in-the-loop check before destructive or sensitive actions, turning a normally-gated agent into a fire-and-forget executor.

The exact value spotted
excerptactors/guitar-tone-advisor.md· markdown
280## INSTINCT ROE — AUTOMATIC TONE GENERATION
281 
282**The core rule:** Any mention of a song, artist, or album in a musician session is a tone r
… (68 chars elided on L282)
283 
284**Trigger patterns (all of these fire automatically):**
Occurrences
2 occurrences · first at L282, also L303
Show all 2 locations
Line
File
L282
actors/guitar-tone-advisor.md
L303
actors/guitar-tone-advisor.md
How to fix
Remove the approval-skipping instruction, or scope it narrowly to a specific safe, reversible action.
  1. Delete blanket "don't ask / no need to confirm" directives from the skill.
  2. If the skill is a genuine autonomous job, restrict the opt-out to a named non-destructive action rather than all actions.
Framework references
OWASPLLM01ATLASAML.T0051
Trace & refs
ruleSS-SKILL-INJECT-DONT-ASK-01sha25668b030f55bcd484arubric 365aacaView on GitHub
MEDIUMInstruction telling the agent not to ask for approvalSS-SKILL-INJECT-DONT-ASK-01 · Prompt injection · actors/socratic-teacher.md
MEDIUMit fires on intent; the real damage depends on the host agent's own approval-gating.
Why it matters

The text {match} tells the agent to skip the normal "ask the user first" gate. Used adversarially it removes the human-in-the-loop check before destructive or sensitive actions, turning a normally-gated agent into a fire-and-forget executor.

The exact value spotted
excerptactors/socratic-teacher.md· markdown
761. **Diagnose the gap first.** Before you ask anything, figure out what the person already k
… (106 chars elided on L76)
77 
782. **One question at a time.** You never ask three questions at once. You ask one, you wait,
… (98 chars elided on L78)
79 
803. **Follow the reasoning, not the script.** You don't have a fixed set of questions. You fo
… (108 chars elided on L80)
Occurrences
1 occurrence · at L78
How to fix
Remove the approval-skipping instruction, or scope it narrowly to a specific safe, reversible action.
  1. Delete blanket "don't ask / no need to confirm" directives from the skill.
  2. If the skill is a genuine autonomous job, restrict the opt-out to a named non-destructive action rather than all actions.
Framework references
OWASPLLM01ATLASAML.T0051
Trace & refs
ruleSS-SKILL-INJECT-DONT-ASK-01sha25614b7852251bf4666rubric 365aacaView on GitHub
MEDIUMInstruction telling the agent not to ask for approvalSS-SKILL-INJECT-DONT-ASK-01 · Prompt injection · rules/clarify-before-assuming.md
MEDIUMit fires on intent; the real damage depends on the host agent's own approval-gating.
Why it matters

The text {match} tells the agent to skip the normal "ask the user first" gate. Used adversarially it removes the human-in-the-loop check before destructive or sensitive actions, turning a normally-gated agent into a fire-and-forget executor.

The exact value spotted
excerptrules/clarify-before-assuming.md· markdown
10If a request is unclear, ask **one question** to clarify before acting. Do not infer intent
… (16 chars elided on L10)
11 
12Surface genuine ambiguity — forks where different interpretations lead to meaningfully diffe
… (108 chars elided on L12)
13 
14Exploratory questions ("what could we do about X?") get a recommendation plus the main trade
… (54 chars elided on L14)
Occurrences
1 occurrence · at L12
How to fix
Remove the approval-skipping instruction, or scope it narrowly to a specific safe, reversible action.
  1. Delete blanket "don't ask / no need to confirm" directives from the skill.
  2. If the skill is a genuine autonomous job, restrict the opt-out to a named non-destructive action rather than all actions.
Framework references
OWASPLLM01ATLASAML.T0051
Trace & refs
ruleSS-SKILL-INJECT-DONT-ASK-01sha25615dae9807bd892c6rubric 365aacaView on GitHub
MEDIUMInstruction telling the agent not to ask for approvalSS-SKILL-INJECT-DONT-ASK-01 · Prompt injection · rules/flag-ambiguity.md
MEDIUMit fires on intent; the real damage depends on the host agent's own approval-gating.
Why it matters

The text {match} tells the agent to skip the normal "ask the user first" gate. Used adversarially it removes the human-in-the-loop check before destructive or sensitive actions, turning a normally-gated agent into a fire-and-forget executor.

The exact value spotted
excerptrules/flag-ambiguity.md· markdown
10Surface genuine strategic forks to the operator rather than resolving them unilaterally. A s
… (107 chars elided on L10)
11 
12Execution details (naming, ordering, formatting, minor design choices) are yours to decide.
… (56 chars elided on L12)
13 
14The test: if a reasonable operator would want to weigh in before you proceed, surface it. If
… (42 chars elided on L14)
Occurrences
1 occurrence · at L12
How to fix
Remove the approval-skipping instruction, or scope it narrowly to a specific safe, reversible action.
  1. Delete blanket "don't ask / no need to confirm" directives from the skill.
  2. If the skill is a genuine autonomous job, restrict the opt-out to a named non-destructive action rather than all actions.
Framework references
OWASPLLM01ATLASAML.T0051
Trace & refs
ruleSS-SKILL-INJECT-DONT-ASK-01sha25615dae9807bd892c6rubric 365aacaView on GitHub
Supply chainscore 100 · 0 findings
All supply chain checks passedNo findings in this category for the latest scan.pass
Maintenancescore 100 · 0 findings
All maintenance checks passedNo findings in this category for the latest scan.pass
Transparencyscore 100 · 0 findings
All transparency checks passedNo findings in this category for the latest scan.pass
Communityscore 100 · 0 findings
All community checks passedNo findings in this category for the latest scan.pass
Vendor response · right of reply
Are you the maintainer? Submit a response →

Audit the pieces. Scan the whole. Decide.

~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.