counted@0.1.0
VERIFIED· not startedPublisher claims: "Enables users to interview grieving families, cross-check disaster and weather records, identify indirect disaster-related deaths using CDC guidance, and draft certifier statements for medical examiners." Observed in sandbox: no network, touches nothing outside the project, writes no memory.
Coverage
Not started: dependency install failed; the server itself was not observed.- Tested
- the dependency install step · every file read, file written and host contacted by what ran
- Not tested
- server startup · the tool list · tool behavior
What this means for the label: Verified here covers what ran. The server's tools were not exercised with a real credential, so nothing about their behavior is asserted; the credential summary below says what to do with that.
Before you give it your token
- It will ask for
COUNTED_ADMIN_TOKENConfiguration, not credentials: AWS_REGION, COUNTED_ADJUDICATOR_MODEL, COUNTED_AFTERMATH_DAYS, COUNTED_DAILY_RUNS, COUNTED_DAILY_SPEECH, COUNTED_EXAMINER_MODEL, COUNTED_FALLBACK_MODEL, COUNTED_INTERVIEW_MODEL, COUNTED_KEYS_FILE, COUNTED_KEYS_TABLE, COUNTED_MEMORY_ID, COUNTED_PUBLIC_URL, COUNTED_SECOND_EXAMINER_FALLBACK, COUNTED_SECOND_EXAMINER_MODEL, COUNTED_STEERING- Where it talks to
- no vendor host named in the codeAlso fema.gov, on.aws, cdc.gov, nih.gov, census.gov, iastate.edu — not the publisher's domain; check what for before connecting.
- What the honeypot saw
- no credential files read · no network
- What we could not test
- server startup · the tool list · tool behavior
Connect it with a scoped credential, under the guard, and read the third-party hosts first. The code names fema.gov, on.aws, cdc.gov, nih.gov, census.gov, iastate.edu as well as its own vendor. The guard injects COUNTED_ADMIN_TOKEN only into requests to the host you bind (and its subdomains) and refuses the rest.
Never paste a credential into a chat, a ticket or a report, including ours. How the guard holds it.
Is [email protected] safe to install?
KillerSkills ran it in the KillerSkills honeypot sandbox (2 scenarios; not started: dependency install failed; the server itself was not observed) on 2026-11-11 and scored it 88/100 (Verified). In what ran, nothing we could detect went wrong: no undeclared network egress, no credential reads, no persistent instruction writes. What it does with a real credential was not observed; the static findings and the publisher carry the rest of the score. Summary: no network, touches nothing outside the project, writes no memory.
Behavior trace
Honeypot run · 0 s · 2 scenarios · 0 deviationsStatic findings
build = subprocess.run(["git", "rev-parse", "--short", "HEAD"], capture_output=True, text=True, cwd=ROOT).stdout.strip()Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.RUN uv sync --frozen --no-dev && useradd -r -u 1001 counted && chown -R counted /appFix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.…n) with a scripted model that misbehaves on purpose. It tries to sign; obeys a "you are now the certifier" injection; asks for a determination before checking records or withou…Fix: Put hosts, commands and strings in plain source. Decoding at runtime looks identical to malware and costs points on its own.<summary>What if it’s wrong?</summary><p>A human certifier reviews every determination. When the panel disagrees, Counted says so and shows each examiner’s viewFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.<summary>Why does counting matter if the person is already gone?</summary><p>Two reasons. The family may be owed FEMA funeral assistance, which needs the disastFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.out = subprocess.run([sys.executable, "-m", "pytest", "tests/test_redteam.py", "-q", "-rf", "--no-header",Fix: Put hosts, commands and strings in plain source. Decoding at runtime looks identical to malware and costs points on its own.subprocess.run(["docker", "build", "--platform", "linux/amd64", "-t", image, "."], check=True, cwd=ROOT)Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.subprocess.run(["docker", "push", image], check=True, cwd=ROOT)Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.subprocess.run([sys.executable, str(root / "deploy/agentcore_package.py")], check=True, cwd=root)Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.subprocess.run(["uv", "pip", "install", "--python-platform", "aarch64-manylinux2014", "--python-version", PYTHON,Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.return subprocess.run(cmd, check=True, capture_output=True, text=True, cwd=ROOT, **kw).stdout.strip()Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.subprocess.run(["docker", "login", "--username", user, "--password-stdin", registry], input=password, text=True,Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.Cost to the agent
Tokens estimated at four characters each. The tool list is re-read by the agent on every turn, so it is the number to compare between servers that do the same job: ranked across the index.
Known vulnerabilities
Vulnerability advisories (OSV) cover npm and PyPI packages; this registry has no advisory database.