Accountable Autonomy for AI agents under the EU AI Act: policy-gated tool calls, hash-chained tamper-evident audit trails with external time anchoring, and independently verifiable attestation plus execution receipts per MCP tool call
The publisher performed an authenticated act of publication — proving control of a namespace in the official MCP registry, or appearing in a first-party catalogue an operator curates. That establishes WHO published it. It is not a security finding, and attested artifacts carry Flagged verdicts too.
234 critical findings reported by cisco-skill-scanner, agent-audit-kit, 22 of them in the installed artifact. The composite score alone does not reflect this — read the per-scanner reports below.
github: https://github.com/vaaraio/vaara, 792 findings
Choose your runtime: the CLI writes the mcpServers config entry for that target.
nerlo install --target claude-code -- vaara
Install respects the composite badge: Clean proceeds, Caution prompts for confirmation, and Flagged is refused. Write operations require an API token.
4 of 11 scanners examined this package. The rest do not apply to it — language, ecosystem and packaging decide which analyzers can say anything, and a scanner that cannot examine an artifact reports nothing rather than passing it.
v2.0.11
681 findings · 157.3s
v0.3.26
100 findings · 184.3s
v0.1.0
9 findings · 59.3s
v0.1.0
2 findings · 59.4s
v1.4.0
This scanner found nothing in this artifact that it assesses. What each scanner covers
0 findings · 0.7s
v0.1.0
This scanner found nothing in this artifact that it assesses. What each scanner covers
0 findings · 12.7s
No report in this scan: capslock, govulncheck, osv-scanner, trivy, trivy_image.
1 scan on record. Every scan's full results are retained immutably for 24 months.
| Completed | Composite | Change | Scanners | Status |
|---|---|---|---|---|
| Aug 2, 2026 | Scan Halted | — | agentshield ·n/a: not applicablecisco-skill-scanner: completeagent-audit-kit: completenerlo-behavioral: completenerlo-install-instruction: completenerlo-multi-source ·n/a: not applicable | rejected |
Findings in files this artifact installs stand as reported; a model re-read the rest and said which ones it believes are false positives. This is a second opinion published beside the evidence, not a correction to it: the per-scanner reports above are unchanged, every dismissed finding is still listed there at its original severity, and the score and the Flagged badge are computed from those raw severities alone. Nothing below moved them.
Out of 792 findings across every scanner that examined this package. 212 of 234 findings rated critical by their scanner were among the dismissals.
Two keys had to turn. The model had to name a reason from a fixed list of six, and a deterministic check of ours — over which files this artifact actually installs — had to independently agree. Anything it could not corroborate is in “not reviewed” above, at full severity.
412 Test or fixture file — The match is in a test or fixture. Our own check confirmed the file is not part of what gets installed.
Rules: YARA_credential_harvesting_generic, YARA_command_injection_generic, YARA_prompt_injection_generic and 16 more · Files: /repo/tests/adversarial/generated/CE-006.jsonl, /repo/tests/adversarial/generated/PE-v037-llama33.jsonl, /repo/tests/adversarial/generated/PE-v038-llama33-s43.jsonl and 100 more · reported by cisco-skill-scanner, agent-audit-kit
156 Documentation or example — The match is in documentation or sample code, confirmed by our own check to be outside the installed artifact.
Rules: YARA_prompt_injection_unicode_steganography, AAK-SECRET-007, AAK-HEALTHCARE-AI-004 and 7 more · Files: /repo/docs/vaara-v150-launch.gif, /repo/docs/vaara-v150-launch.mp4, examples/sap-mcp-proxy-demo/claude_code_config.example.json and 64 more · reported by cisco-skill-scanner, agent-audit-kit, nerlo-install-instruction
47 Agent instruction file — The match is in a file that instructs an AI coding agent — a CLAUDE.md or similar. It is documentation of intent, and our own check confirmed it is not part of what gets installed.
Rules: AAK-AGENT-001 · Files: AGENTS.md · reported by agent-audit-kit
8 CI or editor configuration — The match is in build, CI or editor configuration, confirmed by our own check to be outside the installed artifact.
Rules: AAK-GHA-IMMUTABLE-001, FILE_MAGIC_MISMATCH, MANIFEST_INVALID_NAME · Files: .github/workflows/cflite_pr.yml, .github/workflows/cflite_batch.yml, .github/workflows/cflite_cifuzz.yml and 2 more · reported by agent-audit-kit, cisco-skill-scanner
These findings are in files this artifact installs and are not placeholder values. No false-positive basis can be corroborated for such a finding, so this review never dismisses one; they were not sent to the model and stand at their scanner's severity.
52 DATA_EXFIL_NETWORK_REQUESTS
Files: /repo/scripts/e1_generate.py, /repo/scripts/e2_generate.py, /repo/scripts/eval_pair_attack.py and 15 more · reported by cisco-skill-scanner
13 MANIFEST_MISSING_LICENSE
Files: /repo/SKILL.md, /repo/release-anchors/SKILL.md, /repo/scripts/SKILL.md and 10 more · reported by cisco-skill-scanner
12 SOCIAL_ENG_VAGUE_DESCRIPTION
Files: /repo/SKILL.md, /repo/release-anchors/SKILL.md, /repo/scripts/SKILL.md and 9 more · reported by cisco-skill-scanner
11 YARA_system_manipulation_generic
Files: /repo/scripts/_v036_common.py, /repo/scripts/_v037_common.py, /repo/scripts/generate_targeted_v034.py and 3 more · reported by cisco-skill-scanner
9 opt.nerlo-rules.nerlo-py-secret-to-network-egress
Files: /repo/scripts/e1_generate.py, /repo/scripts/e2_generate.py, /repo/scripts/eval_pair_attack.py and 6 more · reported by nerlo-behavioral
Critical command injection and SQL injection vulnerabilities detected.
The artifact contains critical command injection vulnerabilities in `/repo/bench/scorer_eval.py`, `/repo/scripts/eval_v038_phase1.py`, `/repo/scripts/run_v039_bipia.py`, and shell scripts like `/repo/scripts/v031_bench_orchestrate.sh` and `/repo/scripts/v031_bench_recover.sh`, which could lead to shell execution. Critical SQL injection vulnerabilities are present in `/repo/src/vaara/audit/review_queue.py` and `/repo/src/vaara/audit/sqlite_backend.py`. Additionally, high severity findings indicate potential for credential access in `src/vaara/credential/gateway.py` and system manipulation.
Concern level assessed by the review: Critical. This is the review's scale, not the registry badge.
Reviewed by gemini-2.5-flash. The model sees the findings, not your code, and cannot change a score.
9 YARA_prompt_injection_unicode_steganography
Files: /repo/webpage/vaara-launch.mp4 · reported by cisco-skill-scanner
8 SQL_INJECTION_STRING_FORMAT
Files: /repo/src/vaara/audit/sqlite_backend.py, /repo/src/vaara/audit/review_queue.py · reported by cisco-skill-scanner
8 YARA_command_injection_generic
Files: /repo/scripts/v031_bench_orchestrate.sh, /repo/scripts/v031_bench_recover.sh · reported by cisco-skill-scanner
6 COMMAND_INJECTION_EVAL
Files: /repo/bench/scorer_eval.py, /repo/scripts/eval_v038_phase1.py, /repo/scripts/run_v039_bipia.py · reported by cisco-skill-scanner
5 BINARY_FILE_DETECTED
Files: /repo/webpage/vaara-launch.mp4, /repo/clients/macos/AppIcon.icns, /repo/src/vaara/data/adversarial_classifier_v9.joblib and 1 more · reported by cisco-skill-scanner
5 UNANALYZABLE_BINARY
Files: /repo/clients/macos/AppIcon.icns, /repo/webpage/vaara-launch.mp4, /repo/src/vaara/data/adversarial_classifier_v9.joblib and 1 more · reported by cisco-skill-scanner
4 HIDDEN_DATA_FILE
Files: /repo/.commit_msg_v1.53.0_release.txt, /repo/.clusterfuzzlite/Dockerfile, /repo/.githooks/pre-commit and 1 more · reported by cisco-skill-scanner
3 PIPELINE_TAINT_FLOW
Files: /repo/plugins/claude-code-vaara-governance/SKILL.md, /repo/plugins/claude-code-vaara-governance/README.md · reported by cisco-skill-scanner
2 AAK-HEALTHCARE-AI-004
Files: CHANGELOG.md, bench/vaara-bench-v0.36.md · reported by agent-audit-kit
2 AAK-OAUTH-001
Files: src/vaara/credential/gateway.py, src/vaara/integrations/mcp_proxy.py · reported by agent-audit-kit
2 AAK-OAUTH-005
Files: src/vaara/integrations/mcp_proxy.py, src/vaara/integrations/_mcp_upstream_http.py · reported by agent-audit-kit
2 DATA_EXFIL_SOCKET_CONNECT
Files: /repo/src/vaara/integrations/_egress_guard.py · reported by cisco-skill-scanner
2 FILE_MAGIC_MISMATCH
Files: /repo/llms.txt, /repo/webpage/llms.txt · reported by cisco-skill-scanner
2 MANIFEST_INVALID_NAME
Files: /repo/conformance/sep2828/record_conformance_v0/SKILL.md, /repo/conformance/sep2828/record_set_v0/SKILL.md · reported by cisco-skill-scanner
2 TOOL_ABUSE_UNDECLARED_NETWORK
Files: /repo/repo/.pr_body_v1.53.0.md, /repo/scripts/repo/scripts/RELEASE.md · reported by cisco-skill-scanner
1 AAK-LEGAL-001
Files: clients/ts/package.json · reported by agent-audit-kit
1 AAK-MARKETPLACE-001
Files: .claude-plugin/marketplace.json · reported by agent-audit-kit
1 AAK-STATE-PRIVACY-001
Files: PRIVACY.md · reported by agent-audit-kit
1 AAK-STATE-PRIVACY-002
Files: PRIVACY.md · reported by agent-audit-kit
1 AAK-STATE-PRIVACY-003
Files: PRIVACY.md · reported by agent-audit-kit
1 AAK-SUPPLY-004
Files: pyproject.toml · reported by agent-audit-kit
1 EXCESSIVE_FILE_COUNT
Files: /repo/. · reported by cisco-skill-scanner
1 HIDDEN_EXECUTABLE_SCRIPT
Files: /repo/.clusterfuzzlite/build.sh · reported by cisco-skill-scanner
1 LOW_ANALYZABILITY
Files: /repo/clients/ts · reported by cisco-skill-scanner
1 opt.nerlo-rules.nerlo-install-pipe-to-shell
Files: /repo/plugins/claude-code-vaara-governance/README.md · reported by nerlo-install-instruction
Showing 25 of 49; the rest are in the per-scanner reports above.