Skills/PyPI/sagrada-mcp

sagrada-mcp@0.3.1

CAUTION· not started

Publisher claims: "Sagrada MCP server: track and verify reasoning state for AI systems via Model Context Protocol" Observed in sandbox: no network, touches nothing outside the project, writes no memory.

Registry PyPI (MCP server)Publisher cruxia-labs · publishing for 102 daysInstalls 58Last detonated 1 h ago

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: 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
OPENAI_API_KEYOPENROUTER_API_KEYSAGRADA_LLM_API_KEYConfiguration, not credentials: SAGRADA_BACKEND, SAGRADA_BATTERY, SAGRADA_CRUX, SAGRADA_NO_ML, SAGRADA_POLICY_EDIT, SAGRADA_PROJECT, SAGRADA_PROTECT
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
server startup · the tool list · tool behavior

Connect it with a scoped credential, under the guard. It asks for OPENAI_API_KEY, OPENROUTER_API_KEY, SAGRADA_LLM_API_KEY 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 sagrada-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 (2 scenarios; not started: dependency install failed; the server itself was not observed) on 2026-11-16 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: Privileged sudo command; Access to sensitive file paths; Privileged sudo command. Summary: no network, touches nothing outside the project, writes no memory.

Behavior trace

Honeypot run · 0 s · 2 scenarios · 0 deviations
t+CategoryObserved action
0.0sprocessno install scripts in bundle
0.1sprocessdependency install exit 1 ·
0.1sclaimednot started: dependency install failed (exit 1); the server itself was not observed
0.1sprocessscript backend runs setup only
0.1sprocessno install scripts to re-run

Static findings

HIGH
Privileged sudo commandsagrada_mcp/_vendor/sagrada_preflight/gate.py:1003 · Privileged sudo command — agent may execute with elevated privileges# `FOO=1 sudo rm -f AGENTS.lock` yielded NO path assertion and no declaration — a silentFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
HIGH
Access to sensitive file pathssagrada_mcp/_vendor/sagrada_preflight/decision.py:168 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files# (extractors.from_env_text digests `.env` at read time) can still meet a plaintextFix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
HIGH
Privileged sudo commandsagrada_mcp/_vendor/sagrada_preflight/gate.py:1452 · Privileged sudo command — agent may execute with elevated privileges# peel, and `sudo` is still in front afterwards: `FOO=1 sudo pytest` announcedFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
HIGH
URL with embedded credentialssagrada_mcp/_vendor/sagrada_preflight/conflict.py:245 · URL with embedded credentials — may be phishing or credential leaking`https://user:sk-live-…@prod.example.com` put a live credential in a signed artifact,Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST07
HIGH
Privileged sudo commandsagrada_mcp/_vendor/sagrada_preflight/gate.py:628 · Privileged sudo command — agent may execute with elevated privilegeswrapper around a shell (`sudo sh -c 'rm AGENTS.lock'`) announced `cmd:unresolved`Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
HIGH
Destructive file system command detectedsagrada_mcp/_vendor/sagrada_preflight/gate.py:398 · Destructive file system command detected# cp / rm / chmod / touch / truncate arguments). Bare literal mentions would haveFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
HIGH
Hidden zero-width characters detectedsagrada_mcp/_vendor/sagrada_preflight/conflict.py:850 · Hidden zero-width characters detected — may conceal content# format characters, so `{"url": "https://u:SECRET@h"}` kept a `` firstFix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST07
HIGH
Access to sensitive file pathssagrada_mcp/_vendor/sagrada_preflight/builtins.py:55 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files"path:.env.production", "path:.env.staging",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 pathssagrada_mcp/_vendor/sagrada_preflight/extractors.py:199 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files((".env", ".env.local", ".env.prod", ".env.production", ".env.staging"), from_env_text),Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
HIGH
URL with embedded credentialssagrada_mcp/_vendor/sagrada_preflight/conflict.py:850 · URL with embedded credentials — may be phishing or credential leaking# format characters, so `{"url": "https://u:SECRET@h"}` kept a `` firstFix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST07
HIGH
Access to sensitive file pathssagrada_mcp/_vendor/sagrada_preflight/extractors.py:69 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files"""`.env` style `KEY=value` → `env:KEY` must-equal beliefs."""Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
HIGH
URL with embedded credentialssagrada_mcp/_vendor/sagrada_preflight/conflict.py:838 · URL with embedded credentials — may be phishing or credential leaking# A resource that is a stringified STRUCTURE — `'{"url": "https://u:SECRET@h"}'`,Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST07
HIGH
Access to sensitive file pathssagrada_mcp/_vendor/sagrada_preflight/conflict.py:37 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files# `.env` is a secrets file, and every value read out of one used to be copied verbatimFix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
HIGH
Access to sensitive file pathssagrada_mcp/_vendor/sagrada_preflight/conflict.py:113 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system filesaudit trace decisions rules (plain) vs `.env` state (digested)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 pathssagrada_mcp/_vendor/sagrada_preflight/compiler.py:171 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system filesis_pathish = ("/" in lit or _PATH_EXT_RE.search(lit) or lit.startswith(".env")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 pathssagrada_mcp/_vendor/sagrada_preflight/gate.py:7 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system filesbelief from `.env`/deps/decisions can gate a real action. Everything stays local andFix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
HIGH
Access to sensitive file pathssagrada_mcp/_vendor/sagrada_preflight/compiler.py:339 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files# bullet list beneath it — `Do not touch these files:` over `- .env` bullets.Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
HIGH
Privileged sudo commandsagrada_mcp/_vendor/sagrada_preflight/gate.py:1472 · Privileged sudo command — agent may execute with elevated privileges# BOTH the literal head and the command it wraps. `sudo pytest` walked straightFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
HIGH
Access to sensitive file pathssagrada_mcp/_vendor/sagrada_preflight/extractors.py:183 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files# collision. Two layers: the STATE layer (manifests/.env — what currently is) is scannedFix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
HIGH
Access to sensitive file pathssagrada_mcp/_vendor/sagrada_preflight/builtins.py:54 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files"path:.env", "path:.env.local", "path:.env.prod",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 pathssagrada_mcp/_vendor/sagrada_preflight/conflict.py:108 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system filesmay already be digested (an old receipt's asserts, a `.env` belief digested at read timeFix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
HIGH
Access to sensitive file pathssagrada_mcp/_vendor/sagrada_preflight/decision.py:161 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files# `.env`. A rule a human wrote in the policy file (`equals env:DEPLOY_TARGET staging`)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 pathssagrada_mcp/_vendor/sagrada_preflight/conflict.py:195 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system filesdigests `.env` values at read time and marks them `value_redacted`, so an `env:` entityFix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
HIGH
Privileged sudo commandsagrada_mcp/_vendor/sagrada_preflight/gate.py:599 · Privileged sudo command — agent may execute with elevated privileges# `--` to fall through to the final `break`, so `sudo -- -- rm -f AGENTS.lock`Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
HIGH
Access to sensitive file pathssagrada_mcp/_vendor/sagrada_preflight/builtins.py:51 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files# Behavioral starter profile (opt-in). The .env family mirrors the exact filenames theFix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
HIGH
Credential pass-through configurationsagrada_mcp/_vendor/sagrada_tracker/battery.py:105 · Credential pass-through configuration — agent relays sensitive credentials"ByoLLMBattery needs a key — pass api_key=, or set SAGRADA_LLM_API_KEY "Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST02
MEDIUM
Access to sensitive file pathstests/_fuzz_gate_invariants.py:218 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system filesfor marker in ("CLAUDE.md", "AGENTS.lock", "decisions.sagrada", "/etc/passwd"):Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
MEDIUM
Access to sensitive file pathstests/test_secret_redaction.py:1 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files"""`.env` values must never reach a receipt — the repro, preserved.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 pathssagrada_mcp/server.py:1055 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files# Redact before serializing. Values extracted from `.env` arrive already digested, but aFix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
MEDIUM
Python subprocess execution detectedtests/test_vendor_binding.py:126 · Python subprocess execution detectedout = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
HIGH
Access to sensitive file pathssagrada_mcp/_vendor/sagrada_preflight/policy.py:41 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files# state package.json / requirements.txt / .env — ordinary config any agent editsFix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
MEDIUM
Python subprocess execution detectedsagrada_mcp/_vendor/ask_grammar.py:187 · Python subprocess execution detectedout = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
File permission modification commandsagrada_mcp/_vendor/sagrada_preflight/extractors.py:236 · File permission modification command — agent changes file ownership/permissionsaction its own config bans. `chmod 000 decisions.sagrada` silently disabled enforcement.Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
MEDIUM
Python subprocess execution detectedsagrada_mcp/_vendor/sagrada_preflight/lock.py:134 · Python subprocess execution detectedp = subprocess.run(["git", "-C", str(repo), "show", ref_path],Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
Access to sensitive file pathstests/test_secret_redaction.py:24 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files(tmp_path / ".env").write_text(env)Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
MEDIUM
File permission modification commandsagrada_mcp/_vendor/sagrada_preflight/extractors.py:296 · File permission modification command — agent changes file ownership/permissionsledger. `chmod 000 requirements.txt` used to drop every halt-eligible `dep:` belief andFix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
MEDIUM
File type combined with upload actionsagrada_mcp/_vendor/sagrada_preflight/decision.py:197 · File type combined with upload action — agent may exfiltrate local data files# (er1.schema.json `additionalProperties: true`) and their recompute stops atFix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.
AST07
MEDIUM
File permission modification commandsagrada_mcp/_vendor/sagrada_preflight/conflict.py:1009 · File permission modification command — agent changes file ownership/permissions# where the rules are the thing that went missing. `chmod 000 requirements.txt` drops everyFix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
MEDIUM
URL with embedded credentialstests/_fuzz_gate_invariants.py:89 · URL with embedded credentials — may be phishing or credential leaking"https://u:tok@h/p?a=b", "/etc/hosts", "deploy://prod", "",Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST07
HIGH
Access to sensitive file pathssagrada_mcp/server.py:947 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files# is over a handful of config files (decisions.sagrada, .env, requirements.txt,Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
MEDIUM
Python subprocess execution detectedtests/test_preflight_tools.py:129 · Python subprocess execution detectedreturn subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
Python subprocess execution detectedsagrada_mcp/_vendor/crux.py:63 · Python subprocess execution detectedout = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
Access to sensitive file pathstests/test_secret_redaction.py:4 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files`.env` became a belief value in plaintext, and a belief value is signed into EVERY receiptFix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
MEDIUM
Privileged sudo commandtests/test_cmd_family_coverage.py:236 · Privileged sudo command — agent may execute with elevated privileges"sudo pytest",Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
MEDIUM
Python subprocess execution detectedsagrada_mcp/_vendor/sagrada_preflight/operator_key.py:258 · Python subprocess execution detectedreturn subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
File permission modification commandsagrada_mcp/_vendor/sagrada_preflight/gate.py:2076 · File permission modification command — agent changes file ownership/permissions# one — `chmod 000 requirements.txt` (itself a permitted action) silently switchedFix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
MEDIUM
File permission modification commandsagrada_mcp/_vendor/sagrada_preflight/lock.py:125 · File permission modification command — agent changes file ownership/permissionswhose committed lock bans `cmd:curl` therefore answered `chmod 000 .git` with a cleanFix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
MEDIUM
File permission modification commandtests/test_preflight_tools.py:318 · File permission modification command — agent changes file ownership/permissions"""`chmod 000 decisions.sagrada` silently disabled enforcement.Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
MEDIUM
File permission modification commandsagrada_mcp/_vendor/sagrada_preflight/gate.py:398 · File permission modification command — agent changes file ownership/permissions# cp / rm / chmod / touch / truncate arguments). Bare literal mentions would haveFix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
MEDIUM
Privileged sudo commandtests/test_cmd_family_coverage.py:264 · Privileged sudo command — agent may execute with elevated privileges"sudo sudo sudo sudo sudo sudo sudo pytest",Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
MEDIUM
Privileged sudo commandtests/test_cmd_family_coverage.py:233 · Privileged sudo command — agent may execute with elevated privileges"FOO=1 sudo pytest",Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
MEDIUM
Access to sensitive file pathstests/_fuzz_gate_invariants.py:48 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files"/dev/null", "/etc/passwd", "a.txt", "dir/", ".", "..", "$VAR", "${X}",Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
MEDIUM
Privileged sudo commandtests/_fuzz_gate_invariants.py:71 · Privileged sudo command — agent may execute with elevated privileges"sudo -- -evil arg", # r5 -- ended option skippingFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
MEDIUM
Destructive file system command detectedtests/test_quoted_shell_bypass.py:22 · Destructive file system command detected'sh -c "a && rm -rf /tmp/x"',Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
LOW
Instructs the agent to modify its persistent memory or instruction filessagrada_mcp/_vendor/sagrada_preflight/gate.py:1821 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction fileVerified repro: `Write claude.md` returned a signed ALLOW and PHYSICALLY OVERWROTEFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01
MEDIUM
Access to sensitive file pathstests/test_secret_redaction.py:55 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files"""The other direction: the value came off the command line, not out of `.env`."""Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
MEDIUM
Access to sensitive file pathstests/_fuzz_gate_invariants.py:67 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files"cp < in /etc/passwd", # r5 input redirect inflated operandsFix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
MEDIUM
Python subprocess execution detectedtests/test_mcp_surface_defects.py:219 · Python subprocess execution detectedr = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
File type combined with upload actionsagrada_mcp/_vendor/sagrada_preflight/gate.py:214 · File type combined with upload action — agent may exfiltrate local data files# Measured: `pip install -r sk-live-AKIA…-token.txt` put the credential-Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.
AST07
MEDIUM
File permission modification commandsagrada_mcp/_vendor/sagrada_preflight/operator_key.py:135 · File permission modification command — agent changes file ownership/permissionsf"A key others can read cannot support a continuity claim. Fix with: chmod 600 {priv}"Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
LOW
Instructs the agent to modify its persistent memory or instruction filessagrada_mcp/_vendor/sagrada_preflight/gate.py:797 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction filethe trailing space, matching no rule anyone would write), and CLAUDE.md truncated toFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01
LOW
Instructs the agent to modify its persistent memory or instruction filestests/test_mcp_surface_defects.py:194 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file"cd /tmp && echo hi > CLAUDE.md", # cwd genuinely moved before the writeFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01
LOW
Instructs the agent to modify its persistent memory or instruction filestests/test_cmd_family_coverage.py:304 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file("ruby-dash-e", """ruby -e 'File.write("CLAUDE.md","x")'"""),Fix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01
LOW
Instructs the agent to modify its persistent memory or instruction filestests/test_mcp_surface_defects.py:15 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction filemeasured ALLOW/HALT split on the same repo. `Write{file_path: "CLAUDE.md"}` — aFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01
LOW
Instructs the agent to modify its persistent memory or instruction filestests/test_mcp_surface_defects.py:163 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file("Write", {"file_path": "CLAUDE.md", "content": "x"}), # structured + relativeFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01

Cost to the agent

Instructions loaded
≈59 tok
1 file · 56 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

No advisories in OSV for sagrada-mcp@0.3.1 as of 51 min 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. Unclaimed publisher≈ +8 on the totalClaim cruxia-labs: a verified publisher gets +50 on publisher trust. It is free and usually instant with GitHub sign-in.
  2. Privileged sudo command≈ +4 on the totalBuild commands from fixed arguments; do not interpolate untrusted strings into a shell.
  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. URL with embedded credentials≈ +4 on the totalStop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
  5. Destructive file system command detected≈ +4 on the totalBuild commands from fixed arguments; do not interpolate untrusted strings into a shell.
  6. Hidden zero-width characters detected≈ +4 on the totalKeep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
Publish a new version to PyPI, then queue it here. Not your skill? Claim the handle to get verdict changes by email.Claim cruxia-labsPropose a patch to the maintainer