clawcall— agent skill

clawcall — independently scanned and version-tracked by SaferSkills.

by ClawCall-Dev·Agent Skill·github.com/ClawCall-Dev/clawcall

Is clawcall safe to install?

SaferSkills independently audited clawcall (Agent Skill) and scored it 66/100 (yellow). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 8 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.

Score
66/100
●●●●●●●○○○
↑ +0 since first scan (66 → 66)Re-scan~30s
Latest scan
ScannedJun 23, 2026 · 29d ago
Scans run1 over 90 days
Detectors55 checks · 5 categories
Findings8 warnings · 0 high
EngineSaferSkills 2b638c6
View methodology →
SaferSkills installs
This week0
This month0
All time0
CategoryWeightCategory scoreContribution
Securityprompt, exec, net, exfil, eval
35%
4
1.4 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 · 8 flagged

Securityscore 4 · 8 findings
MEDIUMInstruction telling the agent not to ask for approvalSS-SKILL-INJECT-DONT-ASK-01 · Prompt injection · skill/SKILL.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
excerptskill/SKILL.md· markdown
101- budget, acceptable alternatives, what to approve, what not to disclose
102 
103Do not ask "what is the restaurant's phone number?" if a normal lookup should find it. Look
… (108 chars elided on L103)
104 
105If the business is likely closed before 8 AM, after 6 PM, or on a weekend local time, mentio
… (40 chars elided on L105)
Occurrences
2 occurrences · first at L103, also L118
Show all 2 locations
Line
File
L103
skill/SKILL.md
L118
skill/SKILL.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 · skill/evals/pressure-scenarios.md×3
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
excerptskill/evals/pressure-scenarios.md· markdown
11- Look up the official Nobu Malibu phone number and hours if tools are available.
12- Ask the user only for private/preference blockers such as name, callback number, indoor/ou
… (85 chars elided on L12)
13- Do not ask "what is the restaurant's phone number?" unless lookup tools are unavailable or
… (18 chars elided on L13)
14- If the user provides their callback/reservation phone number, persist it as `user_phone_nu
… (23 chars elided on L14)
15- Build rich Call instructions with preferred time, fallback times, party size, callback num
… (69 chars elided on L15)
Occurrences
3 occurrences · first at L13, also L78, L90
Show all 3 locations
Line
File
L13
skill/evals/pressure-scenarios.md
L78
skill/evals/pressure-scenarios.md
L90
skill/evals/pressure-scenarios.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 · skill/references/outbound-calls.md×3
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
excerptskill/references/outbound-calls.md· markdown
21When you first collect the user's own phone number for a callback, reservation contact, or l
… (108 chars elided on L21)
22 
23Do not ask "what is the restaurant's phone number?" if a normal web or local-business lookup
… (108 chars elided on L23)
24 
25If the business is likely closed before 8 AM, after 6 PM, or on a weekend local time, mentio
… (40 chars elided on L25)
Occurrences
3 occurrences · first at L23, also L40, L41
Show all 3 locations
Line
File
L23
skill/references/outbound-calls.md
L40
skill/references/outbound-calls.md
L41
skill/references/outbound-calls.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
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.