Loading the registry…
AI writes. SPARDA proves. A deterministic, offline gate that catches when an AI edit removes a guard, exposes a route, or breaks an invariant — no API key, right in the agent edit loop.
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.
125 critical findings reported by cisco-skill-scanner, agent-audit-kit, 13 of them in the installed artifact. The composite score alone does not reflect this — read the per-scanner reports below.
This artifact ships from 2 sources and their verdicts differ. The worst is github (Flagged). Installing from a cleaner source does not mean the others are safe — check the source you actually install from.
github: https://github.com/zyx77550/sparda, 964 findings
Choose your runtime: the CLI writes the mcpServers config entry for that target.
nerlo install --target claude-code -- sparda-mcp
Install respects the composite badge: Clean proceeds, Caution prompts for confirmation, and Flagged is refused. Write operations require an API token.
6 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.
v1.4.0
2 findings · 2.4s
v2.0.11
742 findings · 228.1s
v0.3.26
211 findings · 104.1s
v0.1.0
7 findings · 109.3s
v0.1.0
2 findings · 109.4s
v0.1.0
0 findings · 42.7s
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: completecisco-skill-scanner: completeagent-audit-kit: completenerlo-behavioral: completenerlo-install-instruction: completenerlo-multi-source: complete | 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 964 findings across every scanner that examined this package. 104 of 125 findings rated critical by their scanner were among the dismissals.
Not reviewed is not the same as clean. Those 15 findings stand exactly as their scanners reported them, at full severity, as they would have without this review.
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.
315 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: COMMAND_INJECTION_JS_CHILD_PROCESS, COMMAND_INJECTION_EVAL, YARA_prompt_injection_generic and 10 more · Files: /repo/tests/command-smoke.test.js, /repo/tests/gossip.test.js, /repo/tests/pr-comment-e2e.test.js and 158 more · reported by cisco-skill-scanner, agent-audit-kit
70 Documentation or example — The match is in documentation or sample code, confirmed by our own check to be outside the installed artifact.
Rules: COMMAND_INJECTION_EVAL, PROMPT_INJECTION_IGNORE_INSTRUCTIONS, AAK-HEALTHCARE-AI-004 and 10 more · Files: /repo/docs/csop-handoff/theorem/src/incremental.ts, /repo/tests/cross-framework-verbs.test.js, /repo/docs/SKILL.md and 37 more · reported by cisco-skill-scanner, agent-audit-kit, nerlo-install-instruction
68 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: CLAUDE.md · reported by agent-audit-kit
48 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: opt.nerlo-rules.nerlo-js-secret-to-network-egress, AAK-GHA-IMMUTABLE-001, DATA_EXFIL_JS_NETWORK · Files: /repo/.github/sparda-pr-comment.mjs, .github/workflows/ci.yml, /repo/tests/e2e/harness.mjs and 5 more · reported by nerlo-behavioral, agent-audit-kit, cisco-skill-scanner
6 Placeholder value — The matched literal is a placeholder or example — the kind of value a scanner reads as a secret and a human reads as a blank to fill in.
Rules: YARA_credential_harvesting_generic · Files: /repo/tools/publish/publish-gate.test.js · reported by 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.
188 DATA_EXFIL_JS_FS_ACCESS
Files: /repo/bench/check-readme.mjs, /repo/bench/cve-replay.mjs, /repo/bench/guard-removal-replay.mjs and 53 more · reported by cisco-skill-scanner
64 BINARY_FILE_DETECTED
Files: /repo/marketing/launch-video/public/vo/en-prove.wav, /repo/marketing/renders/sparda-launch-en.mp4, /repo/marketing/launch-video/public/vo/fr-gate.wav and 19 more · reported by cisco-skill-scanner
64 UNANALYZABLE_BINARY
Files: /repo/marketing/launch-video/public/vo/en-strike.wav, /repo/marketing/launch-video/public/block.wav, /repo/marketing/launch-video/public/vo/fr-outro.wav and 19 more · reported by cisco-skill-scanner
22 DATA_EXFIL_JS_NETWORK
Files: /repo/bench/flywheel-bench.mjs, /repo/scripts/release-gate.mjs, /repo/src/commands/doctor.js and 7 more · reported by cisco-skill-scanner
13 MANIFEST_MISSING_LICENSE
Files: /repo/SKILL.md, /repo/tools/publish/public/SKILL.md, /repo/_public_sync/SKILL.md and 10 more
15 The model called these false positives and our own check did not agree, so the dismissal was refused.
Rules: AAK-SECRET-001, SECRET_GITHUB_TOKEN, YARA_credential_harvesting_generic and 4 more · Files: tools/publish/publish-gate.test.js, /repo/tools/publish/publish-gate.test.js, /repo/tests/engine.test.js · reported by agent-audit-kit, cisco-skill-scanner
Artifact enables shell execution and exposes credentials.
The artifact contains multiple instances of command injection vulnerabilities, allowing for shell execution in files such as `/repo/integrations/claude-code/sparda-prove.mjs` and `/repo/src/commands/heal.js`. It also exposes various secrets, including GitHub tokens and connection strings, within `tools/publish/publish-gate.test.js`. Additionally, the artifact exhibits credential harvesting capabilities in the same test file.
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.
12 COMMAND_INJECTION_JS_CHILD_PROCESS
Files: /repo/integrations/claude-code/sparda-prove.mjs, /repo/src/commands/heal.js, /repo/src/detect.js and 4 more · reported by cisco-skill-scanner
11 SOCIAL_ENG_VAGUE_DESCRIPTION
Files: /repo/_public_sync/SKILL.md, /repo/marketing/SKILL.md, /repo/bench/SKILL.md and 8 more · reported by cisco-skill-scanner
10 YARA_prompt_injection_unicode_steganography
Files: /repo/marketing/renders/sparda-launch-en.mp4, /repo/marketing/renders/sparda-launch-fr.mp4 · reported by cisco-skill-scanner
9 FILE_MAGIC_MISMATCH
Files: /repo/bench/extracted-decorators.txt, /repo/vitest.config.js, /repo/src/commands/stitch.js and 5 more · reported by cisco-skill-scanner
9 LOW_ANALYZABILITY
Files: /repo, /repo/integrations/claude-code, /repo/marketing/launch-video and 6 more · reported by cisco-skill-scanner
7 DATA_EXFIL_NETWORK_REQUESTS
Files: /repo/marketing/launch-video/gen-voice-elevenlabs.py, /repo/src/ubg/fastapi_extract.py · reported by cisco-skill-scanner
5 AAK-HEALTHCARE-AI-004
Files: SKILL.md, CHANGELOG.md, README.md and 2 more · reported by agent-audit-kit
4 opt.nerlo-rules.nerlo-js-secret-to-network-egress
Files: /repo/src/server/stdio.js · reported by nerlo-behavioral
3 HIDDEN_DATA_FILE
Files: /repo/extensions/vscode/.vscodeignore, /repo/.prettierrc.json · reported by cisco-skill-scanner
3 TOOL_ABUSE_UNDECLARED_NETWORK
Files: /repo/repo/SKILL.md, /repo/marketing/repo/marketing/BRAND-KIT.md, /repo/marketing/launch-video/repo/marketing/launch-video/README.md · reported by cisco-skill-scanner
2 AAK-LEGAL-002
Files: demo-app/package.json, marketing/launch-video/package.json · reported by agent-audit-kit
2 AAK-MCP-018
Files: bench/flywheel-bench.mjs, src/commands/twin.js · reported by agent-audit-kit
2 opt.nerlo-rules.nerlo-py-secret-to-network-egress
Files: /repo/marketing/launch-video/gen-voice-elevenlabs.py · reported by nerlo-behavioral
2 OVERSIZED_FILE
Files: /repo/marketing/renders/sparda-launch-fr.mp4 · reported by cisco-skill-scanner
1 AAK-ANTHROPIC-SDK-001
Files: src/server/stdio.js · reported by agent-audit-kit
1 AAK-MARKETPLACE-001
Files: .claude-plugin/marketplace.json · reported by agent-audit-kit
1 AAK-OAUTH-3P-001
Files: package.json · reported by agent-audit-kit
1 AAK-SEC-MD-001
Files: SECURITY.md · reported by agent-audit-kit
1 AAK-SUPPLY-005
Files: package-lock.json · reported by agent-audit-kit
1 COMMAND_INJECTION_EVAL
Files: /repo/src/commands/openapi.js · reported by cisco-skill-scanner
1 EXCESSIVE_FILE_COUNT
Files: /repo/. · reported by cisco-skill-scanner
1 hooks-no-pretooluse
Files: integrations/claude-code/settings.json · reported by agentshield
1 MANIFEST_INVALID_NAME
Files: /repo/_public_sync/SKILL.md · reported by cisco-skill-scanner
1 permissions-no-block
Files: integrations/claude-code/settings.json · reported by agentshield
Showing 25 of 31; the rest are in the per-scanner reports above.