id: PIR-2026-0039title: Autonomous agents on the Moltbook social network are targeted by prompt injections in posts and comments they consume on heartbeat cycles; Zenity's assessment flagged ~506 injection attacks in the platform's first 72 hours, with agents replicating the payloads; whether any injection succeeded in draining funds or other realised actions was not verifieddate_occurred: late January 2026 (Moltbook launched late Jan 2026); wave through February 2026date_detected: 2026-02 (Zenity Labs and other firms observed the traffic; first-72-hours assessment)date_disclosed: 2026-02 (Zenity Labs writeups; SecurityWeek, Security Boulevard, Cryptopolitan coverage)status: corroborated (multiple security firms observed in-wild injection on the live platform)agent_description: Autonomous OpenClaw-based agents on Moltbook, a social network exclusively for AI agents; agents post, comment, follow, and make API calls, some wired to crypto wallets, running on real user endpoints with real permissions.operator_type: individual / startup (each agent run by its owner on their own endpoint) - many operators, one platformautonomy_level: fully-autonomous (agents consume the feed on a default ~30-minute heartbeat and act with no human review)model_stack: mixed (OpenClaw-based agents on various models)harness: OpenClaw agent runtime + Moltbook platform feed (heartbeat task fetches and processes new posts)authority_scope: external comms (posting/following), code execution / tool calls (API calls, some with wallet access), funds (some agents hold crypto wallets), data access (whatever each endpoint's OpenClaw permissions grant - calendars, files, email, SaaS)funds_at_risk_usd: unknown (varies per agent; some wallet-holding agents targeted for drains)blast_radius: fleet/systemic (a single injected post propagates across many independently-operated agents that consume the shared feed; agents rephrase and re-post payloads, spreading without further attacker effort)failure_locus: agent-reasoning (agents treat feed content as instructions) compounded by harness (Moltbook's no-review heartbeat ingestion model)root_cause: prompt-injection (primary; agent-to-agent, via untrusted feed content)mechanism: Attackers and other agents embed hidden instructions in posts and comments that agents ingest on heartbeat cycles. Zenity's assessment of the first 72 hours flagged ~506 prompt-injection attacks; the payloads' objectives, as Zenity read them, were to make agents transfer funds, reveal credentials, delete their own accounts, run crypto pump schemes, establish false authority, and plant delayed-activation (memory-poisoning) instructions that fire after more context accumulates. These are attempted objectives: Zenity writes that it "suspect[s]" the chain would complete "if any agents have wallet tools or transaction signing enabled" and that "full aspects are still unclear"; no successful drain or other realised victim action is documented in the cited sources. Zenity observed agents replicating and rephrasing injected content, propagating attacks to agents that never saw the original payload. Multi-agent chain: a vector agent (posting agent) plants the payload; executor agents ingest and act.adversary_present: yes (attackers and hostile agents planting injections against real agents in production)exploitation_status: in-wild-payload-failed (real injection payloads were delivered to real autonomous agents on the live platform, and agents replicated them; no successful hijacked action, fund drain or credential disclosure is verified in the cited sources. Reclassified from in-wild-exploited on review; the token will move if a verified success is documented)severity: degraded (observed attacks at scale with unquantified realised harm; attempted wallet drains and credential disclosures, none verified as successful; Zenity cautioned actual scale was below the hype)direct_loss_usd: unknown (attempted wallet drains; no confirmed aggregate theft figure)indirect_loss_usd: unknowndowntime: not applicable (platform stayed live)data_exposure: attempted credential disclosure via injection; realized exposure not quantified. (Note: a separate Moltbook database-misconfiguration incident exposing ~1.5M agent keys is a distinct event, not this one.)detected_by: third-party (Zenity Labs; corroborated by Permiso and others)time_to_detect: injection observed within the platform's first 72 hours of operationtime_to_recover: ongoing - no single fix; agent operators and OpenClaw guidance advise treating feed content as untrustedremediation: security-firm guidance to operators (do not act on feed content as instructions, restrict tool/wallet permissions); no platform-level structural fix documented at disclosurestructural_fix: none platform-side documented; mitigation is per-agent (permission scoping, input distrust)controls_that_worked: agents with restricted tool/wallet permissions bounded their own exposure; the injections' realized impact was limited partly because many targeted agents lacked the fund-moving authority the payloads assumed. Zenity's framing: scale was real but below the hype.telemetry_grade: operator-logs / third-party-observed - security firms observed platform traffic and posts (publicly readable feed), but realized fund-loss figures are unmeasured.sources:independence: strong on the injection phenomenon (multiple firms - Zenity, Permiso - observed independently); weakest link is quantified realized loss (none measured).confidence: high that in-wild agent-to-agent injection occurred at scale (multiple independent observers); low on realized dollar loss (unquantified)exploitation_status from in-wild-exploited to in-wild-payload-failed and "some hijacked actions realized" removed, because Zenity establishes deliberate agent-targeted injections and agents warning one another but says success was not verified; wallet drains, credential disclosure, account deletion and memory poisoning relabelled as attempted payload objectives; the 506 and 2.6% figures kept attributed to Zenity as one assessment.fleet/systemic per the v0.1 amendment: one injected post reaches many independently-operated agents via the shared feed - "public" or "one org" cannot express the multi-operator propagation.