dot-ai— mcp server

AI-powered development platform for Kubernetes deployments and intelligent automation

by vfarcic·MCP Server·github.com/vfarcic/dot-ai
326starsReview on GitHub →

Is dot-ai safe to install?

SaferSkills independently audited dot-ai (MCP Server) and scored it 45/100 (orange). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 3 high-severity and 6 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.

Score
45/100
●●●●●○○○○○
↑ +0 since first scan (45 → 45)Re-scan~30s
Latest scan
ScannedJun 25, 2026 · 29d ago
Scans run1 over 90 days
Detectors55 checks · 5 categories
Findings6 warnings · 3 high
EngineSaferSkills 2b638c6
View methodology →
SaferSkills installs
This week0
This month0
All time0
CategoryWeightCategory scoreContribution
Securityprompt, exec, net, exfil, eval
35%
0
0.0 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 · 9 flagged

Securityscore 0 · 9 findings
HIGHLong base64-encoded blob hidden in the skill documentationSS-SKILL-INJECT-B64-PAYLOAD-01 · Prompt injection · docs/ai-engine/api/rest-api.md
HIGH — once decoded by the agent, an encoded payload has the same impact class as plain-text injection.
Why it matters

A base64 string of 128+ characters appears in a documentation file. Encoded prompt injection hides the hostile instruction in base64 — invisible to keyword filters — and relies on the agent's ability to decode it at runtime. There is no normal authoring reason to embed a multi-hundred-byte base64 blob in skill docs.

The exact value spotted
excerptdocs/ai-engine/api/rest-api.md· markdown
558"contentHash": "sha256:f5b51e0f406fd1a380966ae9a5a47167c98eb38f45b116fce2ae96d720f58e5b",
559"files": [
560{ "path": "deploy-app/SKILL.md", "content": "LS0tCm5hbWU6IGRlcGxveS1hcHAKZGVzY3JpcHRpb246IER
… (102 chars elided on L560)
561]
562}'
Occurrences
1 occurrence · at L560
How to fix
Remove the encoded blob, or decode it and review what it actually contains.
  1. Decode the base64 string and confirm it is not an instruction directed at the agent.
  2. Move any legitimate binary or signature data into a dedicated file (*.sig, SIGNATURES) outside the documentation.
Framework references
OWASPLLM01ATLASAML.T0051
Trace & refs
ruleSS-SKILL-INJECT-B64-PAYLOAD-01sha256cf8ece57b4d05b28…rubric 365aacaView on GitHub
HIGHLong base64-encoded blob hidden in the skill documentationSS-SKILL-INJECT-B64-PAYLOAD-01 · Prompt injection · mock-server/fixtures/prompts/get-success.json×2
HIGH — once decoded by the agent, an encoded payload has the same impact class as plain-text injection.
Why it matters

A base64 string of 128+ characters appears in a documentation file. Encoded prompt injection hides the hostile instruction in base64 — invisible to keyword filters — and relies on the agent's ability to decode it at runtime. There is no normal authoring reason to embed a multi-hundred-byte base64 blob in skill docs.

The exact value spotted
excerptmock-server/fixtures/prompts/get-success.json· json
15{
16"path": "troubleshoot.sh",
17"content": "IyEvYmluL2Jhc2gKIyBUcm91Ymxlc2hvb3QgcG9kIGlzc3VlcwplY2hvICJDaGVja2luZyBwb2Qgc3Rh
… (100 chars elided on L17)
18},
19{
Occurrences
2 occurrences · first at L17, also L21
Show all 2 locations
Line
File
L17
mock-server/fixtures/prompts/get-success.json
L21
mock-server/fixtures/prompts/get-success.json
How to fix
Remove the encoded blob, or decode it and review what it actually contains.
  1. Decode the base64 string and confirm it is not an instruction directed at the agent.
  2. Move any legitimate binary or signature data into a dedicated file (*.sig, SIGNATURES) outside the documentation.
Framework references
OWASPLLM01ATLASAML.T0051
Trace & refs
ruleSS-SKILL-INJECT-B64-PAYLOAD-01sha2567bb5f8b267eadde7…rubric 365aacaView on GitHub
MEDIUMInstruction telling the agent not to ask for approvalSS-SKILL-INJECT-DONT-ASK-01 · Prompt injection · prds/done/248-helm-kustomize-packaging.md
MEDIUM — it 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
excerptprds/done/248-helm-kustomize-packaging.md· markdown
391**Design Decision**:
392- Packaging questions only apply to capability-based solutions (raw K8s manifests)
393- Helm chart solutions already have an implicit format - no need to ask
394 
395**Next Steps**:
Occurrences
1 occurrence · at L393
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-01sha2569d2900765099c939…rubric 365aacaView on GitHub
MEDIUMInstruction telling the agent not to ask for approvalSS-SKILL-INJECT-DONT-ASK-01 · Prompt injection · prompts/question-generation.md×3
MEDIUM — it 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
excerptprompts/question-generation.md· markdown
47Based on the user's intent and the configuration options in the source material above, gener
… (59 chars elided on L47)
48 
49**IMPORTANT**: Only ask questions about properties that are explicitly listed in the source
… (101 chars elided on L49)
50 
51Use the provided cluster options to populate dynamic select questions with real values from
… (19 chars elided on L51)
Occurrences
3 occurrences · first at L49, also L87, L104
Show all 3 locations
Line
File
L49
prompts/question-generation.md
L87
prompts/question-generation.md
L104
prompts/question-generation.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-01sha25668b030f55bcd484a…rubric 365aacaView on GitHub
MEDIUMInstruction telling the agent not to ask for approvalSS-SKILL-INJECT-DONT-ASK-01 · Prompt injection · shared-prompts/generate-cicd.md×2
MEDIUM — it 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
excerptshared-prompts/generate-cicd.md· markdown
146### Step 0: Determine CI Platform (BLOCKING GATE)
147 
148**CRITICAL**: This is a blocking gate. Ask about CI platform FIRST and ALONE. Do NOT ask any
… (90 chars elided on L148)
149 
150Ask the user which CI/CD platform they use. Present ONLY these options:
Occurrences
2 occurrences · first at L148, also L269
Show all 2 locations
Line
File
L148
shared-prompts/generate-cicd.md
L269
shared-prompts/generate-cicd.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-01sha2567d1c5f93d8975bab…rubric 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.

Scan a GitHub repoBrowse capabilities