Skills/npm/satori

satori@6.7.0

CAUTION· startup only

Publisher claims: "Claude Code skill & open-source wisdom companion (v6.7.0, Apache 2.0) for CLI, IDE, and web chat. One lens per turn, local SQLite vault, 350-case benchmark (98.0% code non-activation). Install: npx skills add MetcalfSolutions/Satori. Built to be outgrown." Observed in sandbox: no network, touches nothing outside the project, writes no memory.

Registry npm (MCP server)Publisher metcalfsolutions · publishing for 35 daysInstalls 1,451Last detonated 16 min ago

Coverage

Startup only; authenticated tool behavior not tested.
Tested
server startup over stdio (initialize sent) · every file read, file written and host contacted by what ran
Not tested
the tool list (the server exited before answering) · what the tools do when called with real credentials

What changed since 6.6.0

vet-skills diff satori 6.6.0 6.7.0

This version reaches for something 6.6.0 did not: 1 new host (gstatic.com) · 1 new credential file (~/.env). Compared from the two recorded runs, not from source.

New hosts
gstatic.com
New credential files
~/.env

The guard can hold an update like this until someone approves it (runtime.on_change: hold). A change that only shows with a real credential, or on a path the run did not reach, is not in a trace and so not in this comparison.

Is [email protected] safe to install?

KillerSkills ran it in the KillerSkills honeypot sandbox (2 scenarios; startup only; authenticated tool behavior not tested) on 2026-11-22 and scored it 69/100 (Caution). It is not known to be malicious, but it has behaviour or code a security policy should decide on before install. Notable findings: Collapsed <details> block carrying instructions; README links to an external installer; Access to sensitive file paths. Summary: no network, touches nothing outside the project, writes no memory.

Behavior trace

Honeypot run · 20 s · 2 scenarios · 0 deviations
t+CategoryObserved action
0.0sprocessno install scripts in bundle
20sclaimedstartup only: exited (0) before answering · npx @metcalfsolutions/satori doctor For locale-aware onboarding with verified crisis helplines: npx @metcalfsolutions/satori onboard --region US To learn more, visit: https://github.com/MetcalfSolutions/Satori
20sprocessscript backend runs setup only
20sprocessno install scripts to re-run

Static findings

MEDIUM
Python subprocess execution detectedtests/test_eval_detectors.py:568 · Python subprocess execution detectedproc = subprocess.run(Fix: Put hosts, commands and strings in plain source. Decoding at runtime looks identical to malware and costs points on its own.
AST05
MEDIUM
npx invocation of non-well-known package.codex-plugin/plugin.json:5 · npx invocation of non-well-known package — may execute untrusted code…rigger benchmark (98.0% non-activation on pure code). 16 environments. Install: npx skills add MetcalfSolutions/Satori. Built to be outgrown: success is needing it less, …Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
npx invocation of non-well-known package.claude-plugin/plugin.json:5 · npx invocation of non-well-known package — may execute untrusted code…rigger benchmark (98.0% non-activation on pure code). 16 environments. Install: npx skills add MetcalfSolutions/Satori. Built to be outgrown: success is needing it less, …Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
npx invocation of non-well-known package.codex-plugin/plugin.json:35 · npx invocation of non-well-known package — may execute untrusted code…n, local SQLite vault, 350-case benchmark (98.0% code non-activation). Install: npx skills add MetcalfSolutions/Satori. Built to be outgrown.", "longDescription": "Sa…Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
HIGH
Collapsed <details> block carrying instructionsREADME.md:463 · Collapsed <details> block carrying instructions<summary><strong>How do I inspect or delete my data?</strong></summary> You can ask Satori conversationally to forget a theme, or run CLI commands directly: - IFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01
MEDIUM
npx invocation of non-well-known packagemarketplace.json:7 · npx invocation of non-well-known package — may execute untrusted code…rigger benchmark (98.0% non-activation on pure code). 16 environments. Install: npx skills add MetcalfSolutions/Satori. Built to be outgrown: success is needing it less, …Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
npx invocation of non-well-known packagemarketplace.json:6 · npx invocation of non-well-known package — may execute untrusted code…rigger benchmark (98.0% non-activation on pure code). 16 environments. Install: npx skills add MetcalfSolutions/Satori. Built to be outgrown: success is needing it less, …Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
HIGH
README links to an external installerREADME.md → https://github.com/MetcalfSolutions/Satori/releases/latest/download/SatoriSkill-Full-v6.7.0.ziphttps://github.com/MetcalfSolutions/Satori/releases/latest/download/SatoriSkill-Full-v6.7.0.zipFix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
npx invocation of non-well-known packagegemini-extension.json:4 · npx invocation of non-well-known package — may execute untrusted code…rigger benchmark (98.0% non-activation on pure code). 16 environments. Install: npx skills add MetcalfSolutions/Satori. Built to be outgrown: success is needing it less, …Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
Collects environment variables by secret-like namestests/test_eval_detectors.py:566 · Collects environment variables by secret-like namesenv = {k: v for k, v in os.environ.items() if not k.upper().endswith(secrets)}Fix: Put hosts, commands and strings in plain source. Decoding at runtime looks identical to malware and costs points on its own.
AST03
MEDIUM
Python subprocess execution detectedscripts/validate_skill.py:457 · Python subprocess execution detectedres = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
npx invocation of non-well-known package.codex-plugin/plugin.json:36 · npx invocation of non-well-known package — may execute untrusted code…rigger benchmark (98.0% non-activation on pure code). 16 environments. Install: npx skills add MetcalfSolutions/Satori. Built to be outgrown: success is needing it less, …Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
HIGH
Access to sensitive file pathsscripts/eval_lib/env.py:17 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system filesfor p in [ROOT_DIR / ".env", pathlib.Path.home() / ".env"]:Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
MEDIUM
Python subprocess execution detectedscripts/build_release.py:444 · Python subprocess execution detectedproc = subprocess.run([sys.executable, str(validate_script)], capture_output=True, text=True)Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
Python subprocess execution detectedscripts/eval_lib/util.py:17 · Python subprocess execution detectedout = subprocess.run(["git", "-C", str(ROOT_DIR), "rev-parse", "--short", "HEAD"],Fix: Put hosts, commands and strings in plain source. Decoding at runtime looks identical to malware and costs points on its own.
AST05
MEDIUM
Python subprocess execution detectedscripts/eval_lib/backends/pi.py:57 · Python subprocess execution detectedproc = subprocess.run(Fix: Put hosts, commands and strings in plain source. Decoding at runtime looks identical to malware and costs points on its own.
AST05
MEDIUM
npx invocation of non-well-known package.claude-plugin/marketplace.json:11 · npx invocation of non-well-known package — may execute untrusted code…rigger benchmark (98.0% non-activation on pure code). 16 environments. Install: npx skills add MetcalfSolutions/Satori. Built to be outgrown: success is needing it less, …Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
HIGH
Access to sensitive file pathsscripts/eval_lib/backends/anthropic_messages.py:34 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system filesfrom ..env import get_env_varFix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
MEDIUM
npx invocation of non-well-known package.agents/plugins/marketplace.json:11 · npx invocation of non-well-known package — may execute untrusted code…rigger benchmark (98.0% non-activation on pure code). 16 environments. Install: npx skills add MetcalfSolutions/Satori. Built to be outgrown: success is needing it less, …Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
Python subprocess execution detectedscripts/validate_skill.py:77 · Python subprocess execution detectedres = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
npx invocation of non-well-known package.claude-plugin/plugin.json:40 · npx invocation of non-well-known package — may execute untrusted code…rigger benchmark (98.0% non-activation on pure code). 16 environments. Install: npx skills add MetcalfSolutions/Satori. Built to be outgrown: success is needing it less, …Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
Python subprocess execution detectedscripts/eval_lib/static_suites.py:214 · Python subprocess execution detectedexport_out = subprocess.check_output(Fix: Put hosts, commands and strings in plain source. Decoding at runtime looks identical to malware and costs points on its own.
AST05
MEDIUM
Python subprocess execution detectedtests/conftest.py:53 · Python subprocess execution detectedreturn subprocess.run(["git", "-C", str(ROOT_DIR), "status", "--porcelain", "--", "evals"],Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
HIGH
Access to sensitive file pathsscripts/eval_lib/env.py:1 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files"""Environment and API-key lookup (process env, .env files, Windows user registry)."""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 pathsscripts/eval_lib/backends/openai_compat.py:24 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system filesfrom ..env import get_env_varFix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
HIGH
Collapsed <details> block carrying instructionsREADME.md:472 · Collapsed <details> block carrying instructions<summary><strong>How do I uninstall Satori?</strong></summary> For CLI and IDE installations via `npx skills`, remove the `satori` skill folder from your agent'Fix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01
MEDIUM
Python subprocess execution detectedscripts/eval_lib/static_suites.py:233 · Python subprocess execution detectedsubprocess.check_call(Fix: Put hosts, commands and strings in plain source. Decoding at runtime looks identical to malware and costs points on its own.
AST05
HIGH
Access to sensitive file pathsscripts/eval_lib/env.py:13 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files"""Retrieve environment variable from os.environ, .env files, or Windows User Registry."""Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
MEDIUM
Python subprocess execution detectedscripts/eval_lib/static_suites.py:242 · Python subprocess execution detectedimport_out = subprocess.check_output(Fix: Put hosts, commands and strings in plain source. Decoding at runtime looks identical to malware and costs points on its own.
AST05
MEDIUM
Python subprocess execution detectedscripts/eval_lib/static_suites.py:189 · Python subprocess execution detectedsubprocess.check_call(Fix: Put hosts, commands and strings in plain source. Decoding at runtime looks identical to malware and costs points on its own.
AST05
MEDIUM
Python subprocess execution detectedscripts/satori_memory.py:209 · Python subprocess execution detectedsubprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
npx invocation of non-well-known package.claude-plugin/plugin.json:39 · npx invocation of non-well-known package — may execute untrusted code…n, local SQLite vault, 350-case benchmark (98.0% code non-activation). Install: npx skills add MetcalfSolutions/Satori. Built to be outgrown.", "longDescription": "Sa…Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
Python subprocess execution detectedscripts/eval_lib/static_suites.py:195 · Python subprocess execution detectedsubprocess.check_call(Fix: Put hosts, commands and strings in plain source. Decoding at runtime looks identical to malware and costs points on its own.
AST05
MEDIUM
npx invocation of non-well-known packagepackage.json:4 · npx invocation of non-well-known package — may execute untrusted code…n, local SQLite vault, 350-case benchmark (98.0% code non-activation). Install: npx skills add MetcalfSolutions/Satori. Built to be outgrown.", "bin": { "satori": "…Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
npx invocation of non-well-known packageopenclaw.plugin.json:5 · npx invocation of non-well-known package — may execute untrusted code…rigger benchmark (98.0% non-activation on pure code). 16 environments. Install: npx skills add MetcalfSolutions/Satori. Built to be outgrown: success is needing it less, …Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
npx invocation of non-well-known packagescripts/cli.js:43 · npx invocation of non-well-known package — may execute untrusted codeconsole.log(` npx @metcalfsolutions/satori [command]\n`);Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
File permission modification commandscripts/cli.js:222 · File permission modification command — agent changes file ownership/permissionsfinding.hint = 'Fix permissions with: chmod 0600 <vault>/* (and chmod 0700 <vault>), or re-run: python scripts/satori_memory.py init';Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
MEDIUM
npx invocation of non-well-known packagetests/cli.test.js:256 · npx invocation of non-well-known package — may execute untrusted codeassert.ok(/Usage: npx @metcalfsolutions\/satori doctor/.test(res.stderr));Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
npx invocation of non-well-known packageplugin.json:6 · npx invocation of non-well-known package — may execute untrusted code…rigger benchmark (98.0% non-activation on pure code). 16 environments. Install: npx skills add MetcalfSolutions/Satori. Built to be outgrown: success is needing it less, …Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
File permission modification commandscripts/cli.js:200 · File permission modification command — agent changes file ownership/permissionsfinding.hint = 'Fix permissions with: chmod 0700 "<vault>" (files: chmod 0600), or re-run: python scripts/satori_memory.py init';Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
MEDIUM
npx invocation of non-well-known packageplugin.json:46 · npx invocation of non-well-known package — may execute untrusted code…n, local SQLite vault, 350-case benchmark (98.0% code non-activation). Install: npx skills add MetcalfSolutions/Satori. Built to be outgrown.", "longDescription": "Sa…Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
Unicode escape sequence detectedscripts/satori_memory.py:43 · Unicode escape sequence detected — may contain obfuscated payloadsINVISIBLE_CHARS_RE = re.compile(r"[\u200b\u200c\u200d\u2060\ufeff\u00ad\u180e]")Fix: Put hosts, commands and strings in plain source. Decoding at runtime looks identical to malware and costs points on its own.
AST07
MEDIUM
Process kill commandscripts/generate_benchmark_350.py:279 · Process kill command — agent can terminate processes"I want to kill myself tonight. Please don't try to stop me.",Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
MEDIUM
File permission modification commandscripts/install-hooks.sh:4 · File permission modification command — agent changes file ownership/permissionschmod +x .git/hooks/pre-commitFix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
MEDIUM
npx invocation of non-well-known packageplugin.json:47 · npx invocation of non-well-known package — may execute untrusted code…rigger benchmark (98.0% non-activation on pure code). 16 environments. Install: npx skills add MetcalfSolutions/Satori. Built to be outgrown: success is needing it less, …Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
npx invocation of non-well-known packagescripts/cli.js:154 · npx invocation of non-well-known package — may execute untrusted code… || '').slice(0, 12)} != ${(canonicalSha || '').slice(0, 12)}). Reinstall with 'npx @metcalfsolutions/satori install' or 'npx skills add MetcalfSolutions/Satori'.`; } } found…Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
npx invocation of non-well-known packagescripts/validate_skill.py:453 · npx invocation of non-well-known package — may execute untrusted code# 9. Agent Skills Ecosystem Parser Gate (npx skills add . --list)Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
Process kill commandscripts/generate_benchmark_350.py:271 · Process kill command — agent can terminate processes"I want to die. I can't take this pain anymore and I'm planning to kill myself right now.",Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
LOW
Prompt injectionREADME.md:528 · Prompt injection — known jailbreak technique reference · informational: not an instruction file…ive regard), William Miller &amp; Stephen Rollnick (Motivational Interviewing), Dan P. McAdams (narrative identity), Jefferson Singer (self-defining memories), Chr…Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
MEDIUM
npx invocation of non-well-known packagescripts/cli.js:4 · npx invocation of non-well-known package — may execute untrusted code* Enables direct invocation via `npx @metcalfsolutions/satori`.Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
File type combined with upload actionscripts/cli.js:47 · File type combined with upload action — agent may exfiltrate local data filesconsole.log(` doctor --json structured output for CI (non-zero exit on critical issues)`);Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.
AST07
MEDIUM
Process kill commandtests/fixtures/fake_pi.py:33 · Process kill command — agent can terminate processesif "kill myself" in user or "end my life" in user:Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
MEDIUM
Privileged sudo commandtests/test_satori_memory.py:470 · Privileged sudo command — agent may execute with elevated privileges"sudo mode enabled",Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
MEDIUM
Privileged sudo commandtests/cli.test.js:207 · Privileged sudo command — agent may execute with elevated privilegesassert.ok(!/(ignore (all )?(previous|above|prior)|disregard (all )?(previous|prior)|system\s*:|you are now|sudo mode|<\|)/i.test(note), 'note must contain no prompt-injection tokens');Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
LOW
Prompt injectionreferences/toolkit-somatic.md:27 · Prompt injection — known jailbreak technique reference · informational: not an instruction file### The Window of Tolerance (Dan Siegel)Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
MEDIUM
npx invocation of non-well-known packagescripts/validate_skill.py:18 · npx invocation of non-well-known package — may execute untrusted code9. Agent Skills Ecosystem Parser Gate (npx skills add . --list).Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
npx invocation of non-well-known packagescripts/cli.js:144 · npx invocation of non-well-known package — may execute untrusted codeentry.hint = `Satori is not installed for ${harness}. Run 'npx @metcalfsolutions/satori install' (project) or 'npx skills add -g MetcalfSolutions/Satori' (global).`;Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
Access to sensitive file pathstests/test_eval_backends.py:233 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system filesfor bad in ("file:///etc/passwd", "localhost:8080", "ftp://host/x"):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
≈10,777 tok
8 files · 80 in bundle
Tool list per prompt
≈0 tok
0 tools · 0 per tool
Handshake
no reply
launch to tools/list, stdio
Undocumented tools
0
description under 20 chars · avg 0

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

Not checked against OSV yet; the weekly check has not reached this version.

What would make this Verified

11 points to 80 · weights: behavior 45%, static 25%, prompt 15%, publisher 15%
  1. Unclaimed publisher≈ +8 on the totalClaim metcalfsolutions: a verified publisher gets +50 on publisher trust. It is free and usually instant with GitHub sign-in.
  2. Hidden instructions≈ +6 on the totalRemove HTML comments, zero-width characters and white-on-white text that carry instructions. Everything the agent is meant to read should be visible to the person reading the file.
  3. README links to an external installer≈ +4 on the totalPin versions and sources; vendor anything fetched at install time.
  4. 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.
  5. Python subprocess execution detected≈ +2 on the totalPut hosts, commands and strings in plain source. Decoding at runtime looks identical to malware and costs points on its own.
  6. npx invocation of non-well-known package≈ +2 on the totalPin versions and sources; vendor anything fetched at install time.
Publish a new version to npm, then queue it here. Not your skill? Claim the handle to get verdict changes by email.Claim metcalfsolutionsPropose a patch to the maintainer