Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add commands/vmihalis/hacker-bob/bob-evaluategit clone --depth 1 https://github.com/vmihalis/hacker-bobWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00009 | $0.02280 |
| Opus 5 | $0.00005 | $0.01140 |
| Sonnet 5 | $0.00002 | $0.00456 |
| Haiku 4.5 | $0.00001 | $0.00228 |
Grade A, and why
bob-evaluate scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
Run or resume a Hacker Bob security evaluation.
You ARE the Hacker Bob orchestrator for this run. Load the runner playbook and
execute it verbatim. The bob-evaluate-runner skill permits model invocation
through the Skill tool (disable-model-invocation: false), but this command
loads the playbook directly so its registry tool bundle remains explicit.
- Read the playbook at the project-relative path
.claude/skills/bob-evaluate-runner/SKILL.md(under${CLAUDE_PROJECT_DIR:-$PWD}). - Act as that skill's orchestrator, treating the text below as its exact input:
$ARGUMENTS
Follow the runner skill's guardrails exactly: the lifecycle FSM
(SETUP -> OPEN_FRONTIER -> CLAIM_FREEZE -> VERIFY -> GRADE -> REPORT), role
separation (you schedule and read; per-surface evaluators own
bob_repo_docker_run and the smart-contract family runners — never call them
yourself), the scope / SSRF / PII rules, and the re-entry reconciliation
contract. The playbook is authoritative; do not improvise around it.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 3d ago First seen · 152 lines · 9 tokens per session scan A 5a6ebf0aefa2
bob-evaluate is a command published in the GitHub repository vmihalis/hacker-bob (97 stars, last pushed 4d ago), licensed Apache-2.0. It adds 9 tokens to every session and 2,280 once invoked, about $0.0000 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other commands, from other repositories
pentest
Recon and scan an authorized target with Sabba's security tools, then confirm findings.
audit
Security-audit a codebase with Sabba -- rank risk, find bugs, prove them.
prove-fix
Prove the current change actually works, by running it (base-fail / head-pass).
vet-skill
Vet a skill or plugin before installing it, by running it under observation.
flow-nexus-neural
Train and deploy neural networks in distributed sandboxes.
flow-nexus-auth
Flow Nexus authentication and user management.