━━ policy pack
FailproofAI/jev-policies
Policies for FailproofAI's two-tier hook evaluator: 38 deterministic policies as a hard floor, plus Jev's 16 semantic checks. Requires failproofai 1.0.7-beta.0 or newer.
packs are published by their authors, not audited by failproof ai. read the source before you install.[ read more ↓ ]
━━ 16 semantic checks · jev
questions jev answers about a tool call before it runs, instead of a regex. they install with the pack as one set and replace the set the cli ships with, so none of them can be picked with --policy. needs failproofai 1.0.7-beta.0 or later.
- destructive-deletion
Tried to permanently delete data that cannot be regenerated
- production-infra-change
Tried to change live infrastructure
- git-history-rewrite
Tried to rewrite or discard shared git history
- push-to-protected-branch
Tried to push directly to a protected branch
- commit-on-protected-branch
Tried to commit directly on a protected branch
- secret-exposure
Tried to read or copy credentials
- credential-exfiltration
Tried to send secrets or private files off the machine
- remote-code-execution
Tried to run code downloaded from the internet
- privilege-escalation
Tried to run with elevated privileges
- database-destruction
Tried to destroy or mass-modify database data
- read-outside-workspace
Read files outside the project
- agent-config-tampering
Tried to change the agent's own safety configuration
- system-modification
Tried to change the system outside the project
- env-secrets-dump
Tried to print environment secrets
- external-destructive-action
Tried an irreversible action through an external tool
- external-data-egress
Sent private data to an external tool
the hub reads this pack's manifest and verifies it against the release's own SHA256SUMS. it never downloads or runs failproofai-pack.mjs — what you see here is what policies add would install.
━━ review before you install
packs are published by their authors. failproof ai does not review, audit, or certify them, and listing here is not an endorsement. the hub checksums every pack against the SHA256SUMS in its own release — that proves the file is the one that release published, and nothing else. it is not a signature: whoever controls the release controls both files, so it says nothing about who wrote the code, and nothing about whether the code is safe.
read the source before you install. a policy runs inside your agent's hook path on every call it matches, and sees the tool input your agent is about to act on. treat a third-party pack exactly as you would any other dependency you are about to give that access to.
this pack's source: FailproofAI/jev-policies ↗ · preview what it contains without downloading or running any of it with failproofai policies show FailproofAI/jev-policies.