Can other agents read your agent card?

What the card declares

Name

Sigil

Transport

JSONRPC

Endpoint

https://sigilsec.ai

Card shape

pre-v1.0

Location

standard location

Skills

12

Response time

117 ms

Findings

• ErrorRequired field missing: versionversion
• ErrorRequired field missing: defaultInputModesdefaultInputModes
• ErrorRequired field missing: defaultOutputModesdefaultOutputModes
• ErrorRequired field missing: protocolVersionprotocolVersion
• ErrorRequired field missing: provider.organizationprovider.organization
• ErrorRequired field missing: skills[0].tagsskills[0].tags
• ErrorRequired field missing: skills[1].tagsskills[1].tags
• ErrorRequired field missing: skills[2].tagsskills[2].tags
• ErrorRequired field missing: skills[3].tagsskills[3].tags
• ErrorRequired field missing: skills[4].tagsskills[4].tags
• ErrorRequired field missing: skills[5].tagsskills[5].tags
• ErrorRequired field missing: skills[6].tagsskills[6].tags
• ErrorRequired field missing: skills[7].tagsskills[7].tags
• ErrorRequired field missing: skills[8].tagsskills[8].tags
• ErrorRequired field missing: skills[9].tagsskills[9].tags
• ErrorRequired field missing: skills[10].tagsskills[10].tags
• ErrorRequired field missing: skills[11].tagsskills[11].tags
• WarningCard written in the shape used before v1.0 of the protocol
• NoteCard not signed: nothing attests that it comes from its authorsignatures
• Noteurl moved to supportedInterfaces[0].url in v1.0url

How to fix it

• versionThe version of the agent itself, not of the protocol.

  "version": "1.0.0"

• defaultInputModesThe media types the agent accepts, across all its skills.

  "defaultInputModes": ["text/plain", "application/json"]

• defaultOutputModesThe media types the agent answers with.

  "defaultOutputModes": ["text/plain", "application/json"]

• protocolVersionThe A2A version the agent speaks, in the shape used before v1.0.

  "protocolVersion": "0.3"

• provider.organizationThe name of the organization behind the agent.

  "provider": {"organization": "ACME", "url": "https://acme.example"}

• skills[0].tagsKeywords describing the skill. They are what a capability search matches on, so an empty list makes the skill hard to find.

  "tags": ["cooking", "recipes"]

• skills[1].tagsKeywords describing the skill. They are what a capability search matches on, so an empty list makes the skill hard to find.

  "tags": ["cooking", "recipes"]

• skills[2].tagsKeywords describing the skill. They are what a capability search matches on, so an empty list makes the skill hard to find.

  "tags": ["cooking", "recipes"]

Declared skills

• Scan File or DirectoryScan a local file or directory for supply chain security threats. Runs eight-phase static analysis: install hooks, code patterns, network exfiltration, credentials, obfuscation, provenance, prompt injection, and skill security.
• Scan Package from RegistryDownload and scan a package from npm or PyPI. The package is downloaded into quarantine and scanned across all eight phases before anything is installed.
• Clone and Scan RepositoryClone a git repository into quarantine and scan it. The repo is held in quarantine until explicitly approved or rejected.
• List Quarantined ItemsList all items currently held in quarantine with their status, source, risk score, and quarantine ID.
• Approve Quarantined ItemRelease a quarantined item from quarantine for installation or use. Moves the item out of the quarantine directory.
• Reject Quarantined ItemPermanently delete a quarantined item. The downloaded files are removed from the quarantine directory.
• Grade PathLetter grade (A-F), recommendation, behaviour profile and key risks for a path. Cheaper to read than sigil_scan when only the verdict matters.
• Scan Host for Agent-Tooling ResidueRead-only scan of this machine for what installed agent tooling left behind: shell rc edits, cron/launchd/systemd persistence, git hooks, world-readable credential files, leftover tool directories, /etc/hosts redirects, global agent packages. Nothing is changed.
• Plan Residue CleanupShow the reversible fixes Sigil would make for host residue, without applying them. Apply and rollback are not exposed as tools; a human runs them in a terminal.
• Check Package in Scan DatabaseLook up a package or skill's risk assessment in the Sigil public scan database. Works for ClawHub skills, PyPI packages, npm packages, GitHub repos, and MCP servers. Does not require the Sigil CLI.
• Search Scan DatabaseSearch the Sigil public scan database for packages by name or keyword. Returns scanned packages with their verdicts and risk scores. Does not require the Sigil CLI.
• Report ThreatReport a malicious file to the Sigil threat intelligence database by its SHA256 hash. Requires the Sigil CLI and an authenticated session (sigil login). Reports are reviewed by the security team.

Show the result

This badge follows your readings: it changes when your card changes. Paste it in your README.

Markdown

[![A2A card](https://kopass.app/a2a/badge/aHR0cHM6Ly93d3cuc2lnaWxzZWMuYWk.svg)](https://kopass.app/a2a/validator?url=https://www.sigilsec.ai)

HTML

<a href="https://kopass.app/a2a/validator?url=https://www.sigilsec.ai"><img src="https://kopass.app/a2a/badge/aHR0cHM6Ly93d3cuc2lnaWxzZWMuYWk.svg" alt="A2A card" height="20"></a>