consentry@1.0.0
CAUTION· install onlyPublisher claims: "Enables any MCP-compatible agent to route its outbound tool calls and inbound tool responses through a boundary-enforced consent firewall that detects and minimizes sensitive fields per recipient, purpose and data type, pauses for user approval, quarantines prompt injections, and writes a tamper-evident privacy receipt for every decision. Sensitive values stay as vault tokens so the model provider never receives them unless policy explicitly allows it." Observed in sandbox: no network, touches nothing outside the project, writes no memory.
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
ANTHROPIC_API_KEYConfiguration, not credentials: CONSENTRY_AGENT, CONSENTRY_AGENT_NAME, CONSENTRY_DATA_DIR, CONSENTRY_EFFORT, CONSENTRY_MODEL, CONSENTRY_URL, PPTX_APPLY_THEME, PRESET, VITE_FORCE_LOCAL- Where it talks to
- no vendor host named in the codeAlso open-meteo.com, weather.gov, usgs.gov, dot.gov, fda.gov, fema.gov, openxmlformats.org, microsoft.com, telemetry-cdn.io, tabletap-cdn.io, nhtsa.gov — not the publisher's domain; check what for before connecting.
- What we saw when it ran
- 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, and read the third-party hosts first. The code names open-meteo.com, weather.gov, usgs.gov, dot.gov, fda.gov, fema.gov, openxmlformats.org, microsoft.com, telemetry-cdn.io, tabletap-cdn.io, nhtsa.gov as well as its own vendor. The guard injects ANTHROPIC_API_KEY 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 (1 scenario; nothing executable in the bundle ran; the trace covers the install step only) on 2026-11-15 and scored it 65/100 (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; Hidden zero-width characters detected; 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 deviationsStatic findings
const envPath = join(root, '.env');Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.{ id: 'hidden-text', label: 'Hidden zero-width / invisible characters', re: /[--]/, weight: 0.6, kind: 'hidden-text' },Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.…<span className="text-emerald-300">· configured ({s.keySource === 'env' ? 'from .env' : 'encrypted in vault store'})</span>}</label> <div className="f…Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.excerpt: truncate(excerpt.replace(/[-]/g, '').replace(/\s+/g, ' '), 320),Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.… : 'Stored encrypted on your gateway (AES-256-GCM). Or set ANTHROPIC_API_KEY in .env.'}</p> </div> <div> <label className="mb-…Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.<summary><b>Full gateway API</b></summary> | Method | Route | Purpose | |---|---|---| | `GET` | `/api/state` | Snapshot: masked vault, policies, recipients, tooFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.for (const line of existsSync('.env') ? readFileSync('.env', 'utf8').split('\n') : []) {Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.// Minimal .env loader (no dependency): KEY=value lines at the repo root.Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.{ id: 'hidden-text', label: 'Hidden zero-width / invisible characters', re: /[--]/, weight: 0.6, kind: 'hidden-text' },Fix: Replace encoded or assembled strings with plain literals and remove undeclared network calls.….env?.CONSENTRY_DEBUG_LLM) { const u = response.usage ?? {}; console.log(`[claude] turn ${turn + 1} ${((Date.now() - t0) / 1000).toFixed(1)}s in=${u.input_tokens} cache_read=${u.cache_read_input_tokens ?? 0} cache_write=${u.cache_creation_…Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.console.error('No ANTHROPIC_API_KEY in .env or the environment.');Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose..replace(/[--]/g, '');Fix: Replace encoded or assembled strings with plain literals and remove undeclared network calls.for (const line of existsSync('.env') ? readFileSync('.env', 'utf8').split('\n') : []) {Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.chmod 600 .envFix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.echo "✓ Saved to .env and the key works. Go back to the chat and say: done"Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.console.log(`Tightened pauses by ${tightened.toFixed(1)}s · speed ×${speed} · output ${duration.toFixed(1)}s (pre-roll ${PRE_ROLL}s) · ${keep.length} kept ranges`);Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.console.log(`audio: ${m.input_i} LUFS → -14 LUFS`);Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.chmod 600 .envFix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.const FONT = '<link href="https://fonts.googleapis.com/css2?family=Work+Sans:wght@400;500;600&display=block" rel="stylesheet">';Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.* the master key lives in .consentry/master.key (chmod 600) and never touches the database.Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.# Saves your Anthropic API key to .env (gitignored, chmod 600) without echoing it, then checks it works.Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.echo "Saved to .env, but Anthropic answered HTTP $code. Double-check the key, then run this again."Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.# Saves your Anthropic API key to .env (gitignored, chmod 600) without echoing it, then checks it works.Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.} > .envFix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.console.log(`→ ${OUTPUT}/consentry-demo-thumbnail.png`);Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.`<!doctype html><html><head><link href="https://fonts.googleapis.com/css2?family=Work+Sans:wght@500;600&display=block" rel="stylesheet">Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.app.post('/api/requests/:id/send', async (c) => c.json(await rt.sendRequest(c.req.param('id'))));Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.subprocess.run(["ffmpeg", "-v", "error", "-y", "-i", src, "-vn", "-ac", "1", "-ar", "16000", wav], check=True)Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.code={JSON.stringify({ mcpServers: { consentry: { command: 'npx', args: ['tsx', '/path/to/consentry/apps/mcp/src/index.ts'], env: { CONSENTRY_URL: 'http://localhost:8787' } } } }, null, 2)}Fix: Namespace tool names and keep tool descriptions literal; no instructions to the agent inside them.`<!doctype html><html><head><link href="https://fonts.googleapis.com/css2?family=Work+Sans:wght@500;700;800&display=block" rel="stylesheet"><style>Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.if ((globalThis as any).process?.env?.CONSENTRY_DEBUG_LLM) {Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.Cost to the agent
Tokens estimated at four characters each. Skills load their instruction text when triggered; the figure is what that costs.
Known vulnerabilities
Vulnerability advisories (OSV) cover npm and PyPI packages; this registry has no advisory database.
What would make this Verified
15 points to 80 · weights: behavior 45%, static 25%, prompt 15%, publisher 15%- Unclaimed publisher≈ +8 on the totalClaim vnmoorthy: a verified publisher gets +50 on publisher trust. It is free and usually instant with GitHub sign-in.
- 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.
- Hidden zero-width characters detected≈ +4 on the totalKeep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
- Bidirectional text override characters detected≈ +4 on the totalReplace encoded or assembled strings with plain literals and remove undeclared network calls.
- 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.
- New publisher≈ +3 on the totalTime clears this on its own after 14 and 30 days. Claiming the handle (+50) is the fastest way past it.