PUBLIC REPORT · ADDED BY METAL MANTRA / AGENT SIGNAL v0.2

cumora.

Scanned repository says: Where agent teams gather.

Metal Mantra scanned this public repository on its own initiative; the maintainer did not submit it. Maintainers: scan the repository yourself to take this report over, or contact us.

Inspect GitHub source

3.9k starsMITpushed 6d ago

0community interest

Votes show community interest, not safety. They never change this score.

AUTOMATED SIGNAL61/100CCCNot security certification

START HERE

Review findings need attention.

LLM request without token ceiling

Inspect the findings

WHAT THIS DOES NOT ESTABLISH

Runtime safety, real-world agent capability, fitness for your use case or independent certification.

Scan scope199 / 356Partial eligible-file selection · not the entire repository

Recorded findings15Heuristic matches requiring human review

SnapshotOct 12, 2026UTC · engine 0.4.0

DNS at scan timeUnverifiedDomain control is not security certification

01 / KNOW THE REPOSITORY

What the source
actually tells us.

Structured, cited information from the scanned archive. Repository-authored claims are labeled and never treated as verified capabilities.

INSPECTED SOURCE200 / 356Partial selection

READMEReadRepository-authored information

MANIFESTS2LICENSE, package.json

Who it is for

Stated purpose · repository says · unreviewed
Where agent teams gather.
README.md:3
Stated purpose · repository says · unreviewed
cumora.ai · Web app · Desktop download · iOS beta (TestFlight)
README.md:5
Stated purpose · repository says · unreviewed
Two "brain" paths:
README.md:22

What it can do

No explicit capability or integration was extracted; that does not mean none exists.

What adoption takes

Stated prerequisite · manifest declares · unreviewed
Manifest declares Node.js >=20 as a requirement.
package.json:13
Declared dependency · manifest declares · unreviewed
Declares a dependency on openai.
package.json:95
Declared dependency · manifest declares · unreviewed
Declares a dependency on @aws-sdk/client-s3.
package.json:57
Declared dependency · manifest declares · unreviewed
Declares a dependency on @aws-sdk/s3-request-presigner.
package.json:58
Declared dependency · manifest declares · unreviewed
Declares a dependency on @capacitor/android.
package.json:59
Declared dependency · manifest declares · unreviewed
Declares a dependency on @capacitor/app.
package.json:60
Declared dependency · manifest declares · unreviewed
Declares a dependency on @capacitor/browser.
package.json:61

Limits and terms

Declared license · manifest declares · unreviewed
Manifest declares MIT as its license.
package.json:11

Observed technical context

TypeScript · 199 selected filesJavaScript · 1 selected fileModel API reference · static onlyTool-calling reference · static only

These are references in selected source files, not proof that a feature works at runtime.

Read from commit 6224983 at Oct 12, 2026, 5:08 AM UTC. Source links open that exact commit. Repository statements are unreviewed and do not affect the score.

02 / FINDINGS FIRST

What needs
a closer look.

Critical details are withheld publicly for the first 30 days after a scan. A matched pattern is a lead for investigation, not a confirmed vulnerability.

Dependencies: 564 exact runtime versions from package-lock.json checked against OSV.dev on Oct 12, 2026; 19 with known advisories. Reachability is not checked.

SeverityPillar15 findings

MAJOR

LLM request without token ceiling

REL001 · Guardrails & loop control

A model request does not set a maximum output length. One unusual input can produce a very long, expensive response.

SUGGESTED ACTION

Set max_tokens or the provider's equivalent on each model call.

server/src/novita.ts:275

MAJOR

LLM request without token ceiling

REL001 · Guardrails & loop control

A model request does not set a maximum output length. One unusual input can produce a very long, expensive response.

SUGGESTED ACTION

Set max_tokens or the provider's equivalent on each model call.

server/src/novita.ts:317

MAJOR

Sensitive action without human approval

REL004 · Guardrails & loop control

The code can delete data, send messages or move money, and no step asks a person to approve it first. A wrong model decision becomes a real action.

SUGGESTED ACTION

Require explicit human approval before a destructive or external action.

src/components/EmailComposer.tsx:360

MAJOR

LLM output without runtime schema

SCH001 · Schema & tooling

Model output is used without being checked against a schema. Unexpected or malformed output can break later steps or be trusted when it should not be.

SUGGESTED ACTION

Validate model output with Zod, JSON Schema, Pydantic or provider structured output.

server/src/agents/agenda.ts:483

MAJOR

LLM output without runtime schema

SCH001 · Schema & tooling

Model output is used without being checked against a schema. Unexpected or malformed output can break later steps or be trusted when it should not be.

SUGGESTED ACTION

Validate model output with Zod, JSON Schema, Pydantic or provider structured output.

server/src/agents/convene.ts:288

MAJOR

Model call without fallback

EFF001 · Token & cost efficiency

A model call has no visible fallback. If the provider fails or is rate limited, the whole agent fails with it.

SUGGESTED ACTION

Configure a secondary model or explicit fallback path for provider failures.

server/src/agents/agenda.ts:483

MAJOR

Domain ownership not verified

TRUST001 · Entity trust

Metal Mantra could not confirm that the owner controls the stated domain. The domain may belong to someone else.

SUGGESTED ACTION

Verify the domain using the Metal Mantra DNS TXT challenge.

cumora.ai · location unavailable

MAJOR

Dependency with a high-severity advisory

DEP001 · Security & privacy

@capacitor/android 8.3.3 (npm) has a known advisory: GHSA-rvm3-566m-v7fv (highest severity critical). Capacitor Android and iOS: remote content can be loaded at the app origin via the internal HTTP proxy path

SUGGESTED ACTION

Upgrade to a fixed version listed in the advisory and regenerate the lockfile.

package-lock.json:1439

MAJOR

Dependency with a high-severity advisory

DEP001 · Security & privacy

@capacitor/ios 8.3.3 (npm) has a known advisory: GHSA-rvm3-566m-v7fv (highest severity critical). Capacitor Android and iOS: remote content can be loaded at the app origin via the internal HTTP proxy path

SUGGESTED ACTION

Upgrade to a fixed version listed in the advisory and regenerate the lockfile.

package-lock.json:1688

MAJOR

Dependency with a high-severity advisory

DEP001 · Security & privacy

tar 7.5.15 (npm) has known advisories: GHSA-23hp-3jrh-7fpw, GHSA-8x88-c5mf-7j5w, GHSA-gvwx-54wh-qm9j, GHSA-r292-9mhp-454m, GHSA-vmf3-w455-68vh (highest severity critical). node-tar: Decompression/parse DoS via unlimited input

SUGGESTED ACTION

Upgrade to a fixed version listed in the advisory and regenerate the lockfile.

package-lock.json:1636

MINOR

Privacy Policy link not confirmed

TRUST003 · Entity trust

No privacy policy link was found on the stated homepage. Buyers cannot see how the vendor handles their data.

SUGGESTED ACTION

Expose a clear Privacy Policy link on the official homepage.

cumora.ai · location unavailable

MINOR

Terms link not confirmed

TRUST004 · Entity trust

No terms link was found on the stated homepage. Buyers cannot see the conditions that apply to using the service.

SUGGESTED ACTION

Expose a clear Terms link on the official homepage.

cumora.ai · location unavailable

MINOR

Dependency with a known advisory

DEP002 · Security & privacy

@opentelemetry/core 2.2.0 (npm) has a known advisory: GHSA-8988-4f7v-96qf (highest severity moderate). OpenTelemetry Core: Unbounded memory allocation in W3C Baggage propagation

SUGGESTED ACTION

Upgrade to a fixed version listed in the advisory and regenerate the lockfile.

package-lock.json:3657

MINOR

Dependency with a known advisory

DEP002 · Security & privacy

@opentelemetry/core 2.7.1 (npm) has a known advisory: GHSA-8988-4f7v-96qf (highest severity moderate). OpenTelemetry Core: Unbounded memory allocation in W3C Baggage propagation

SUGGESTED ACTION

Upgrade to a fixed version listed in the advisory and regenerate the lockfile.

package-lock.json:3760

MINOR

Dependency with a known advisory

DEP002 · Security & privacy

body-parser 2.2.2 (npm) has a known advisory: GHSA-v422-hmwv-36x6 (highest severity low). body-parser vulnerable to denial of service when invalid limit value silently disables size enforcement

SUGGESTED ACTION

Upgrade to a fixed version listed in the advisory and regenerate the lockfile.

package-lock.json:6662

03 / EVIDENCE & LIMITS

The boundary
is part of the result.

What was scanned: 199 of 356 eligible files · Partial selection.
Supported static source patterns and domain checks; no runtime execution.
What remains unknown: Unscanned files, external dependencies, production configuration, actual agent behavior and evidence beyond the configured checks.
Source location: Links open the current repository HEAD. Source may have changed since this snapshot.
Something looks wrong?: Tell us if a finding, scope or claim on this report needs a correction. Report a problem with this report
Entity trust · 15% of v0.2: Includes DNS ownership, HTTPS and policy evidence. Verification is domain control, not a security certificate.
DNS: unverifiedHTTPS: availablePrivacy: not confirmedTerms: not confirmed

04 / THE METHOD BEHIND THE NUMBER

Open the
calculation.

Each pillar starts at 100. Rule deductions reduce its score; weights determine the overall contribution.

Security & privacy

6 findings · 25% weight

58/10014.5 pts

Guardrails & loop control

3 findings · 25% weight

66/10016.5 pts

Schema & tooling

2 findings · 20% weight

68/10013.6 pts

Token & cost efficiency

1 findings · 15% weight

86/10012.9 pts

Entity trust

3 findings · 15% weight

25/1003.8 pts

Inspect the current standard

USER REPORTS

Did cumora work for people?

No user reports yet.

Each report is one signed-in GitHub account saying what happened. We do not verify that the account ran the agent, and reports never change the automated score or grade. For a security problem, use the disclosure page instead of a public note.

cumora.

Review findings need attention.

What the sourceactually tells us.

Who it is for

What it can do

What adoption takes

Limits and terms

Observed technical context

What needsa closer look.

LLM request without token ceiling

LLM request without token ceiling

Sensitive action without human approval

LLM output without runtime schema

LLM output without runtime schema

Model call without fallback

Domain ownership not verified

Dependency with a high-severity advisory

Dependency with a high-severity advisory

Dependency with a high-severity advisory

Privacy Policy link not confirmed

Terms link not confirmed

Dependency with a known advisory

Dependency with a known advisory

Dependency with a known advisory

The boundaryis part of the result.

Open thecalculation.

Security & privacy

Guardrails & loop control

Schema & tooling

Token & cost efficiency

Entity trust

Did cumora work for people?

A signal.Not a seal.

What the source
actually tells us.

What needs
a closer look.

The boundary
is part of the result.

Open the
calculation.

A signal.
Not a seal.