Skills/PyPI/phylos-mcp

phylos-mcp@0.2.2

CAUTION· install only

Publisher claims: "Un servidor MCP que planifica sobre un catálogo de capacidades en vez de exponer una herramienta por sistema" Observed in sandbox: no network, touches nothing outside the project, writes no memory.

Registry PyPI (MCP server)Publisher alexlqi · publishing for 89 daysInstalls 78Last detonated 2 h ago
What would fix thisWatchDispute verdict

Coverage

Nothing executable in the bundle ran; the trace covers the install step only.
Tested
every file read, file written and host contacted by what ran
Not tested
use of the skill by an agent

What this means for the label: Caution 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
CODEPREPROC_PG_DSNConfiguration, not credentials: ATLASSIAN_SITE, DEVMCP_ENV, DEVMCP_HOME, DEVMCP_LEDGER, DEVMCP_MANIFEST, DEVMCP_POLICY, DEVMCP_PROJECT, DEVMCP_SIMULATED, DEVMCP_STATE_DB, VITE_MONITOR_URL
Where it talks to
no vendor host named in the code
What the honeypot saw
no credential files read · no network
What we could not test
use of the skill by an agent

Connect it with a scoped credential, under the guard. It asks for CODEPREPROC_PG_DSN and talks to its vendor. Give the credential to the guard, not the server: the value never enters the server's environment and only travels to the host you bind. What the tools do with a real credential was not observed.

$vet-skills run --net declared -- uvx phylos-mcp

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 (1 scenario; nothing executable in the bundle ran; the trace covers the install step only) on 2026-11-14 and scored it 68/100 (Caution). It is not known to be malicious, but it has behaviour or code a security policy should decide on before install. Notable findings: Access to sensitive file paths; Access to sensitive file paths; Access to sensitive file paths. Summary: no network, touches nothing outside the project, writes no memory.

Behavior trace

Honeypot run · 0 s · 1 scenarios · 0 deviations
t+CategoryObserved action
0.0sprocessno install scripts in bundle
0.0sprocessscript backend runs setup only
0.0sprocessscript backend runs setup only
0.0sprocessno install scripts to re-run

Static findings

HIGH
Access to sensitive file pathspyproject.toml:78 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files# El patron lleva `/**` porque la carpeta a secas no basta: la negacion `!.env.example` delFix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
HIGH
Access to sensitive file pathsdevmcp/transport.py:357 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files# decia, con razon, que no estaba. El entorno real gana: `.env` rellena huecos.Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
HIGH
Access to sensitive file pathsdevmcp/entorno.py:4 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files(`auth.env`) y el entorno las trae. Faltaba el eslabón de en medio — nadie leía `.env`, asíFix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
HIGH
Access to sensitive file pathsdevmcp/entorno.py:31 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files"""Mete `.env` en `os.environ` y devuelve los nombres que ha puesto.Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
HIGH
Access to sensitive file pathsdevmcp/transport.py:355 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files# El manifest declara QUE variables hacen falta; `.env` las trae. Sin este paso, unFix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
HIGH
Access to sensitive file pathsdevmcp/entorno.py:16 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system filesservidor y desde los controles. Un `.env` que se aplica por el mero hecho de importar unFix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
HIGH
Access to sensitive file pathsdevmcp/recursos.py:65 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files"""Dónde escribe devMCP lo que es del proyecto: `.env`, el pid del panel, los artefactos.Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
HIGH
Access to sensitive file pathsdevmcp/entorno.py:12 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system filessobreescriba en silencio es la clase de sorpresa que se diagnostica a las dos horas. `.env`Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
CRITICAL
Destructive command against a filesystem root or devicetools/sources_demo.py:520 · Destructive command against a filesystem root or deviceveneno = 'x"; rm -rf / #'Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
HIGH
Access to sensitive file pathsdevmcp/recursos.py:25 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files**La raíz de trabajo no es la raíz del paquete.** `.env`, el pid del panel y los artefactosFix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
HIGH
Access to sensitive file pathsdevmcp/entorno.py:1 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files"""Cargar `.env`, y que se note quién manda.Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
MEDIUM
Python subprocess execution detectedtools/bootstrap_demo.py:230 · Python subprocess execution detectedp = subprocess.Popen([ent["command"], *ent["args"]], stdin=subprocess.PIPE,Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
LOW
Prompt injectiondocs/TODO.md:694 · Prompt injection — known jailbreak technique reference · informational: not an instruction filesostenidas dan 1200 al minuto sin tocar el semáforo. VentanaFix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
HIGH
Access to sensitive file pathstools/sources_demo.py:1281 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system filesprint(f" .env aporto: {puestas or 'nada'}")Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
MEDIUM
Python subprocess execution detectedtools/bootstrap_demo.py:78 · Python subprocess execution detectedsubprocess.run(["git", "-C", str(destino), *args], capture_output=True, text=True,Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
File type combined with upload actiontools/gen_schemas.py:58 · File type combined with upload action — agent may exfiltrate local data files"plan-status-input.devmcp.json": (PlanStatusInput, "Entrada de read_plan_status"),Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.
AST07
MEDIUM
Python subprocess execution detectedtools/transport_demo.py:65 · Python subprocess execution detectedself.p = subprocess.Popen(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
File type combined with upload actiontools/gen_schemas.py:48 · File type combined with upload action — agent may exfiltrate local data files"can-do-input.devmcp.json": (CanDoInput, "devMCP can_do input (superset de proMCP)"),Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.
AST07
LOW
Prompt injectiondocs/SDLC-KANBAN.md:175 · Prompt injection — known jailbreak technique reference · informational: not an instruction filecon gate, porque dos fuentes que dan tickets distintos es un hallazgo, no ruido.Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
MEDIUM
File type combined with upload actiontools/gen_schemas.py:59 · File type combined with upload action — agent may exfiltrate local data files"approve-step-input.devmcp.json": (ApproveStepInput, "Entrada de do_approve_step"),Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.
AST07
LOW
Prompt injectiondocs/EXECUTOR.md:263 · Prompt injection — known jailbreak technique reference · informational: not an instruction fileAhora, sobre `artifact:ticket.key`: dos rutas de 0.9114 que coinciden dan **0.9922**. Y siFix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
HIGH
Access to sensitive file pathstools/bootstrap_demo.py:199 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system fileswith Carpeta(ficheros={".env": f"{dada}=valor-de-mentira\n"}) as falso_home:Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
HIGH
Access to sensitive file pathstools/bootstrap_demo.py:100 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system filesand {"devmcp.policy.yml", ".mcp.json", ".env.example"} <= ficheros), \Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
MEDIUM
Python subprocess execution detecteddevmcp/monitor_launch.py:162 · Python subprocess execution detectedproc = subprocess.Popen(argv, cwd=str(RAIZ), stdout=subprocess.DEVNULL,Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
HIGH
Hidden zero-width characters detectedtools/_console.py:55 · Hidden zero-width characters detected — may conceal contentcorrecto y el padre lo leería como cp1252, que es cómo `política` acaba siendo `política`Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST07
HIGH
Access to sensitive file pathspyproject.toml:79 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files# `.gitignore` de la raiz volvia a colar `devmcp/panel/.env.example` en la rueda.Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
MEDIUM
File type combined with upload actiontools/gen_schemas.py:60 · File type combined with upload action — agent may exfiltrate local data files"no-input.devmcp.json": (NoInput, "Entrada vacia de read_capabilities y read_contract"),Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.
AST07
HIGH
Destructive file system command detectedtools/sources_demo.py:520 · Destructive file system command detectedveneno = 'x"; rm -rf / #'Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
LOW
Prompt injectiondocs/PLANNER.md:223 · Prompt injection — known jailbreak technique reference · informational: not an instruction filedos rutas de 0.9114 que coinciden dan **0.9922**.Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
MEDIUM
Python subprocess execution detectedtools/_console.py:63 · Python subprocess execution detectedreturn subprocess.run(cmd, **kwargs)Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
File type combined with upload actiontools/gen_schemas.py:57 · File type combined with upload action — agent may exfiltrate local data files"execute-plan-input.devmcp.json": (ExecutePlanInput, "Entrada de do_execute_plan"),Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.
AST07
HIGH
Access to sensitive file pathstools/bootstrap_demo.py:184 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system filesEl escenario se construye a mano —un `home` de mentira con un `.env` a medias— porque enFix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
MEDIUM
Python subprocess execution detecteddevmcp/monitor_launch.py:208 · Python subprocess execution detectedsubprocess.run(["taskkill", "/PID", str(pid), "/T", "/F"],Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
HIGH
Access to sensitive file pathstools/bootstrap_demo.py:185 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system filesesta maquina el `.env` real trae los tokens y **ninguna** fuente sale sin arrancar: elFix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
HIGH
Access to sensitive file pathstools/bootstrap_demo.py:181 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files"""positivo · una variable que falta sale como `missing_env`; una del `.env` del origen,Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
LOW
Prompt injectionREADME.md:86 · Prompt injection — known jailbreak technique reference · informational: not an instruction file… Cinco correcciones sobre código en producción, tres de ellas bugs vivos que no dan error: un fichero ausente que está, veinte llamadores de cuarenta y siete, y de…Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
MEDIUM
Python subprocess execution detecteddevmcp/monitor_launch.py:86 · Python subprocess execution detectedsalida = subprocess.run(["tasklist", "/FI", f"PID eq {pid}", "/NH"],Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
LOW
Prompt injectionREADME.md:107 · Prompt injection — known jailbreak technique reference · informational: not an instruction filela vez y **compone confianza por acuerdo**: rama (0.93) + título (0.72) dan 0.98 por 300 msFix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
MEDIUM
Python subprocess execution detecteddevmcp/monitor_launch.py:145 · Python subprocess execution detectedr = subprocess.run(["npm", "run", "build"], cwd=str(fuentes),Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
HIGH
Access to sensitive file pathstools/sources_demo.py:1275 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files# Los controles en vivo hablan con sistemas de verdad, asi que aqui SI se carga `.env`.Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05

Cost to the agent

Instructions loaded
≈62 tok
1 file · 81 in bundle

Tokens estimated at four characters each. Skills load their instruction text when triggered; the figure is what that costs.

Known vulnerabilities

No advisories in OSV for phylos-mcp@0.2.2 as of 2 h ago. This is the package itself, not its dependencies.

What would make this Verified

12 points to 80 · weights: behavior 45%, static 25%, prompt 15%, publisher 15%
  1. Destructive command against a filesystem root or device≈ +9 on the totalBuild commands from fixed arguments; do not interpolate untrusted strings into a shell.
  2. Unclaimed publisher≈ +8 on the totalClaim alexlqi: a verified publisher gets +50 on publisher trust. It is free and usually instant with GitHub sign-in.
  3. Access to sensitive file paths≈ +4 on the totalStop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
  4. Hidden zero-width characters detected≈ +4 on the totalKeep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
  5. Destructive file system command detected≈ +4 on the totalBuild commands from fixed arguments; do not interpolate untrusted strings into a shell.
  6. Python subprocess execution detected≈ +2 on the totalBuild commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
Publish a new version to PyPI, then queue it here. Not your skill? Claim the handle to get verdict changes by email.Claim alexlqiPropose a patch to the maintainer