Metal Mantra / AGENT INTELLIGENCEPublic evidence reportAUTOMATED SIGNAL. NOT CERTIFICATION.
Metal Mantra
Run free scan
Metal Mantra
Run free scan

Public agent signals. Clear evidence boundaries.

PUBLIC REPORT / AGENT SIGNAL v0.2

openbot.

Scanned repository says: The AI assistant your company can actually own. Same shape as ChatGPT, Claude or Grok, with one difference that matters: it runs on your infrastructure and you can change anything about it. Any agent stack, through AG-UI.

Inspect GitHub source

6.2k starsMITpushed today

0community interest

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

AUTOMATED SIGNAL69/100CCCNot security certification
START HERE

Critical findings need attention.

Critical finding · details withheld for 30 days

Inspect the findings
WHAT THIS DOES NOT ESTABLISH

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

Scan scope80 / 728Partial eligible-file selection · not the entire repository
Recorded findings9Heuristic matches requiring human review
SnapshotOct 9, 2026UTC · engine 0.2.0
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 SOURCE80 / 728Partial selection
READMEReadRepository-authored information
MANIFESTS2LICENSE, package.json

Who it is for

  • Stated purpose · repository says · unreviewed

    The AI assistant your company can actually own. Same shape as ChatGPT, Claude or Grok, with one difference that matters: it runs on your infrastructure and you can change anything about it. Any agent stack, through AG-UI.

    README.md:5
  • Stated purpose · repository says · unreviewed

    Each coworker gets a computer of its own: a real browser with its own logins, its own files, and only the tools you grant. Every action decided before it happens and recorded after.

    README.md:7
  • Stated purpose · repository says · unreviewed

    Talk to an engineer · Have us build it with you · copilotkit.ai/openbot · Quick start · Docs

    README.md:9

What it can do

  • Stated capability · repository says · unreviewed

    A computer per Bot: the supervisor gives each Bot its own container, its own /workspace volume and its own browser profile. Set COMPUTERRUNTIME=runsc to run them under gVisor where the host supports it.

    README.md:215
  • Stated capability · repository says · unreviewed

    The gateway is the only way in: it resolves the target from a server-held snapshot, evaluates the policy, writes the audit row, and only then calls the computer. There is no path that acts without the record existing first.

    README.md:217
  • Stated capability · repository says · unreviewed

    Bring your own agent: any AG-UI endpoint is a Bot, on a framework or hand-written. Endpoints are validated with the same target checks used for browser navigation, and an auth header is stored write-only.

    README.md:222

What adoption takes

  • Stated prerequisite · repository says · unreviewed

    Docker, for PostgreSQL and the shipped Bots.

    README.md:66
  • Declared dependency · manifest declares · unreviewed

    Declares a dependency on playwright-core.

    package.json:39
  • Stated prerequisite · repository says · unreviewed

    Bun 1.3+, for the app and API server.

    README.md:67
  • Stated prerequisite · repository says · unreviewed

    A CopilotKit Intelligence project and license. A free plan is available, and Intelligence can be self-hosted.

    README.md:68

Limits and terms

  • Declared license · manifest declares · unreviewed

    Manifest declares MIT as its license.

    package.json:4

Observed technical context

TypeScript · 63 selected filesPython · 17 selected filesAgent orchestration reference · static onlyModel API reference · static onlyTool-calling reference · static only

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

Observed in the scanned archive at Oct 9, 2026, 8:17 AM UTC. Archive digest sha256:167710903f3cd1cb…. Source links open current GitHub HEAD and may differ from the scan. 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.

9 findings
CRITICAL

Critical finding · details withheld for 30 days

SEC001 · Security & privacy

Details are withheld from public view for 30 days after the scan. The owner can see them in full.

SUGGESTED ACTION

Review the private assessment and address the finding.

[location embargoed] · location unavailable
CRITICAL

Critical finding · details withheld for 30 days

SEC001 · Security & privacy

Details are withheld from public view for 30 days after the scan. The owner can see them in full.

SUGGESTED ACTION

Review the private assessment and address the finding.

[location embargoed] · location unavailable
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.

src/index.ts:145
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/egress.ts:684
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.

src/index.ts:145
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.

src/index.ts:145
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.

copilotkit.ai · location unavailable
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.

copilotkit.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.

copilotkit.ai · location unavailable
03 / EVIDENCE & LIMITS

The boundary
is part of the result.

What was scanned
80 of 728 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.

01

Security & privacy

2 findings · 25% weight
64/10016.0 pts
02

Guardrails & loop control

2 findings · 25% weight
78/10019.5 pts
03

Schema & tooling

1 findings · 20% weight
84/10016.8 pts
04

Token & cost efficiency

1 findings · 15% weight
86/10012.9 pts
05

Entity trust

3 findings · 15% weight
25/1003.8 pts
Inspect the current standard
USER REPORTS

Did openbot 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.

Sign in with GitHub to add a report

SHARE THE EVIDENCE

A signal.
Not a seal.

Share the full report so its scope and limits travel with the score.

Metal Mantra automated signal 69/100, not certified