Loading the registry…
Design a GPU experiment as a DATA-COLLECTION spec WITH the researcher, then dispatch it to a fresh-context executor. The "together" stage: propose with taste + a recommendation, surface the load-bearing choices, write the DESIGN.md data-collection spec (purpose — what the data is designed to inform; arms; the canonical metric + exact eval definitions; comparability; cost — but NOT a pre-registered verdict), run the pre-launch gates (verify-claim on the FACTS + cross-family design-audit on the DATA-TRUSTABILITY), iterate till the researcher clears it, write the thin START.md executor brief + CHECKLIST.md gates, and dispatch a fresh-context executor to run it. Use when starting to design / scope / propose an experiment ("let's design X", "propose an experiment for Y"), BEFORE it runs. The layer ABOVE run-experiment — that skill EXECUTES the locked brief this one produces.
Claude Skills are markdown + scripts loaded into Claude Code; scanned with the same pipeline.
We found this by crawling public sources and nobody has attested to it. That is how most of the registry gets here, and it is not a mark against the artifact — it only means no authenticated act of publication is on record. An artifact can be the vendor's own and still appear here.
11 critical findings reported by cisco-skill-scanner, agent-audit-kit, 10 of them in the installed artifact. The composite score alone does not reflect this — read the per-scanner reports below.
Claude Skills install to a single target, no runtime picker.
nerlo install -- 'antondelafuente/automated-researcher (design-experiment)'
Manual alternative: copy the skill folder into ~/.claude/skills/ so Claude Code loads it on next launch.
cp -r -- 'antondelafuente/automated-researcher (design-experiment)' ~/.claude/skills/
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
82 findings · 20.3s
v0.3.74
19 findings · 7.3s
v0.1.0
3 findings · 7.2s
v0.1.0
86 findings · 7.2s
v1.4.0
None of the entry points this scanner reads were found in this artifact. What each scanner covers
0 findings · 0.2s
vv0.3.2
No Go packages were found in this artifact. What each scanner covers
0 findings · 0.2s
v0.1.0
This scanner found nothing in this artifact that it assesses. What each scanner covers
0 findings · 0.9s
No report in this scan: 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 22, 2026 | Scan Halted | — | agentshield ·n/a: not applicablecisco-skill-scanner: completeagent-audit-kit: completenerlo-behavioral: completenerlo-install-instruction: completecapslock ·n/a: not applicablenerlo-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 190 findings across every scanner that examined this package. 1 of 11 findings rated critical by their scanner was 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.
6 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-005, AAK-HEALTHCARE-AI-004 · Files: AGENTS.md · reported by agent-audit-kit
6 Documentation or example — The match is in documentation or sample code, confirmed by our own check to be outside the installed artifact.
Rules: SOCIAL_ENG_VAGUE_DESCRIPTION, opt.nerlo-rules.nerlo-install-run-bundled-artifact, MANIFEST_MISSING_LICENSE · Files: /repo/.github/prompts/SKILL.md, /repo/.github/triage/SKILL.md, /repo/.github/prompts/address-review.md and 1 more · reported by cisco-skill-scanner, nerlo-install-instruction
4 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-AGENT-TRUST-002, AAK-AGENT-TRUST-001, FILE_MAGIC_MISMATCH · Files: .github/workflows/review-on-pr.yml, .github/workflows/implement-on-ready.yml, .github/workflows/triage-assess.yml and 1 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.
77 opt.nerlo-rules.nerlo-install-run-bundled-artifact
Files: /repo/CHANGELOG.md, /repo/README.md, /repo/plugins/experiment-lifecycle/skills/design-experiment/SKILL.md and 42 more · reported by nerlo-install-instruction
24 MANIFEST_MISSING_LICENSE
Files: /repo/plugins/verify-claims/skills/verify-claims/SKILL.md, /repo/plugins/experiment-lifecycle/skills/update-site/SKILL.md, /repo/plugins/experiment-lifecycle/skills/run-experiment/SKILL.md and 21 more · reported by cisco-skill-scanner
13 SOCIAL_ENG_VAGUE_DESCRIPTION
Files: /repo/SKILL.md, /repo/proposals/SKILL.md, /repo/tooling/review-eval/SKILL.md and 10 more · reported by cisco-skill-scanner
12 DATA_EXFIL_NETWORK_REQUESTS
Files: /repo/plugins/gpu-job/skills/gpu-job/scripts/deploy_pod.py · reported by cisco-skill-scanner
8 TOOL_ABUSE_SYSTEM_MODIFICATION
Files: /repo/plugins/experiment-lifecycle/skills/update-site/scripts/update_site_smoke.sh, /repo/plugins/experiment-lifecycle/skills/update-dashboard/scripts/update_dashboard_smoke.sh, /repo/plugins/experiment-lifecycle/skills/log-experiment/scripts/log_experiment_secret_scan_smoke.sh
Artifact performs system modification and remote code execution.
The artifact contains scripts that perform system modification in `/repo/plugins/experiment-lifecycle/skills/*/scripts/*.sh`. It also includes scripts that fetch and execute remote code in `/repo/plugins/gpu-job/skills/gpu-job/scripts/teardown.sh`. Additionally, there are instances of secrets being used in conjunction with network egress in `/repo/plugins/gpu-job/skills/gpu-job/scripts/deploy_pod.py` and hidden executable scripts in `/repo/.aar-ci/checks.sh` and related files.
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.
6 opt.nerlo-rules.nerlo-install-shell-inline-command
Files: /repo/plugins/experiment-lifecycle/skills/run-experiment/SKILL.md, /repo/plugins/experiment-lifecycle/skills/run-experiment/scripts/local_job_queue.sh · reported by nerlo-install-instruction
5 AAK-MARKETPLACE-001
Files: .claude-plugin/marketplace.json · reported by agent-audit-kit
4 AAK-HEALTHCARE-AI-004
Files: proposals/111-self-contained-product.md, proposals/387-direct-cli-launch.md, proposals/263-lifecycle-audit-config-gates.md and 1 more · reported by agent-audit-kit
4 FILE_MAGIC_MISMATCH
Files: /repo/plugins/experiment-lifecycle/skills/update-site/scripts/resolve_visualization_recipe.sh, /repo/plugins/experiment-lifecycle/skills/update-dashboard/scripts/resolve_viewer_recipe.sh · reported by cisco-skill-scanner
4 PROMPT_INJECTION_IGNORE_INSTRUCTIONS
Files: /repo/proposals/151-triage-artifact-schema.md, /repo/proposals/SKILL.md · reported by cisco-skill-scanner
3 HIDDEN_EXECUTABLE_SCRIPT
Files: /repo/.aar-ci/checks.sh, /repo/.aar-ci/checks_marketplace_smoke.sh, /repo/.aar-ci/fake_home_smoke.sh · reported by cisco-skill-scanner
3 MANIFEST_DESCRIPTION_TOO_LONG
Files: /repo/plugins/experiment-lifecycle/skills/update-site/SKILL.md, /repo/plugins/verify-claims/skills/verify-claims/SKILL.md, /repo/plugins/experiment-lifecycle/skills/update-dashboard/SKILL.md · reported by cisco-skill-scanner
3 opt.nerlo-rules.nerlo-py-secret-to-network-egress
Files: /repo/plugins/gpu-job/skills/gpu-job/scripts/deploy_pod.py · reported by nerlo-behavioral
2 COMPOUND_FETCH_EXECUTE
Files: /repo/plugins/gpu-job/skills/gpu-job/scripts/teardown.sh · reported by cisco-skill-scanner
2 EXCESSIVE_FILE_COUNT
Files: /repo/proposals/., /repo/. · reported by cisco-skill-scanner
2 HIDDEN_DATA_FILE
Files: /repo/.aar-ci/config, /repo/.githooks/pre-commit · reported by cisco-skill-scanner
1 AAK-SECRET-004
Files: plugins/gpu-job/skills/gpu-job/scripts/teardown.sh · reported by agent-audit-kit
1 opt.nerlo-rules.nerlo-install-pipe-to-shell
Files: /repo/plugins/gpu-job/skills/gpu-job/scripts/bootstrap_pod.sh · reported by nerlo-install-instruction
Showing 25 of 27; the rest are in the per-scanner reports above.