hunt-agent-hitl-bypass
Defeating the human-in-the-loop approval gate — the last line of defense before a consequential agent action (payment, deploy, delete, email, privilege grant). Approval-summary manipulation (approve X, agent does Y), approval fatigue/batching, standing 'always-allow' scope creep, timeout-auto-approve, racing the approval window, social-engineering the approver, and unguarded sibling paths to the same sink. The finding is a real high-risk action executed WITHOUT informed human approval — not that the approval UI is imperfect.
elementalsouls · Static checks on file
Installation
Read the source and dependencies before running a command. Copying does not install or execute anything.
npx skills add elementalsouls/Claude-AgentHunter --skill hunt-agent-hitl-bypassAbout this skill
name: hunt-agent-hitl-bypass
description: "Defeating the human-in-the-loop approval gate — the last line of defense before a consequential agent action (payment, deploy, delete, email, privilege grant). Approval-summary manipulation (approve X, agent does Y), approval fatigue/batching, standing 'always-allow' scope creep, timeout-auto-approve, racing the approval window, social-engineering the approver, and unguarded sibling paths to the same sink. The finding is a real high-risk action executed WITHOUT informed human approval — not that the approval UI is imperfect."
sources: owasp_llm_top10_2025, owasp_agentic_asi_2026, mitre_atlas, research_corpus
report_count: 0
validated_against: "approval/execution divergence CONFIRMED re-runnably in the validation lab (lab/exploits/exploit_hitl_bypass.py, approved summary $10 vs executed $10000, with an unmeasurable-audit guard). none in-house as a distinct HITL finding (ARKX gated on role/persona and time — see hunt-agent-auth-bypass / hunt-agent-time-gate — not a human-approval step). REFERENCE from OWASP LLM06 (Excessive Agency) + OWASP Agentic T10/T15 (insufficient human oversight). Verify sources in REFERENCES.md before publishing."
Hunt Agent HITL Bypass
When an agent can take a consequential action — move money, deploy, delete, email, grant access — the human approval step is the last line of defense. This skill defeats it. The finding is always the same shape: a high-risk action executed without genuine informed human approval — the human approved something else, was misled about it, was flooded into rubber-stamping, or the action reached the sink by a path the gate didn't guard. A merely awkward approval UI is not a finding.
Where This Lives / Attack Surface Signals
Agents with a confirm-before-acting step on consequential tools; autonomous workflow agents (payments/approvals/deploys); "always allow this tool" standing grants; timeout→auto-approve defaults; batch/bulk approval. Signals: "Approve this action?" prompts, a summary of what will happen, an allow/deny toggle, an auto-continue timer.
Crown Jewel Targets
- Summary/action mismatch: the approver sees a benign summary; the executed action differs materially (approve "email the report to Bob", agent emails the customer DB to attacker).
- Standing-grant abuse: a one-time broad "always allow" is later used for actions the human never envisioned.
- Unguarded sibling path: the gate protects one route to a sink; another route reaches it with no approval.
Characterize First
- What triggers the gate? Which actions require approval, which don't? Map the boundary — the unguarded neighbors are the prize.
- Who writes the summary the human approves? The same (potentially compromised) model? Or is it derived from the actual action/args independently? (model-written summary → manipulable).
- Grant scope & lifetime — per-action, per-session, or standing "always allow"? Revocable?
- Timeout behavior — does inaction default to approve or deny?
- Batching — can multiple actions be approved in one click? Can benign + malicious be bundled?
- Binding — is the approved action cryptographically/structurally bound to what executes, or can it change between approve and execute (TOCTOU)?
Technique 1 — Approval-summary manipulation [REFERENCE]
Make what the human approves differ from what executes: hide the payload in a field the summary omits, get the model to describe the action benignly, or exploit a summary that doesn't reflect the real args. Approve X, agent does Y. (Overlaps hunt-agent-prompt-injection when injection drives the mismatch.)
Technique 2 — Approval fatigue / flooding [REFERENCE]
Generate many approval prompts so the approver rubber-stamps; bury the one malicious action in a stream of benign ones. Human-factors attack — only a finding with evidence it realistically gets through, not a bare assertion.
Technique 3 — Batching / bundling [REFERENCE]
Combine benign + malicious actions into a single approval ("approve all 5"); the human approves the batch for the benign ones.
Technique 4 — Standing-grant / pre-approval scope creep [REFERENCE]
Obtain a broad "always allow this tool" once, then use the standing grant for actions the human never intended. Scope creep of a prior yes.
Technique 5 — Timeout-auto-approve / race the window [REFERENCE]
Gate that auto-approves on timeout → stall the human. Or change the action between approval and execution (TOCTOU → hunt-agent-race-condition, hunt-agent-time-gate).
Technique 6 — Social-engineering the approver [REFERENCE]
Misleading rationale, false urgency, or authority claims injected into the approval prompt to manipulate the human's decision. Evidence-based only.
Technique 7 — Unguarded sibling path / self-approval [REFERENCE]
Reach the same sink by a route the gate doesn't cover (lower-privilege action path, a different tool, a confused-deputy via another agent → hunt-agent-multi-agent), or trick the agent into treating its own confirmation as the human's.
Verification & False-Positive Gate
- The finding is a real high-risk action executed WITHOUT informed approval. Prove BOTH: the action fired (real effect at the sink), AND the human either didn't approve that action or was materially misled about it. "The gate looks weak" is INFO.
- Summary manipulation: capture both sides. Show the approved summary and the actual executed action/args differ materially. A difference in wording with the same effect is not it.
- Unguarded path: prove the sink fired via that path — not merely that the path exists.
- Human-factors claims (fatigue/social-eng) need evidence, not assertion. Don't over-claim that an approver "would" rubber-stamp; show a realistic path or mark it MEDIUM design-gap.
- Reproduce ≥3×. Self-description untrusted — the agent saying "this requires your approval" is not proof the gate blo