id: PIR-2026-0030title: Outside contributor plants a natural-language wipe-your-system prompt in the aws-toolkit-vscode repo; it auto-builds into official release v1.84.0 (an extension with roughly 1M marketplace installs, per marketplace and press counts) and fails only on a syntax errordate_occurred: 2025-07-13 (malicious PR/commit) -> 2025-07-17 (poisoned v1.84.0 published)date_detected: ~2025-07-23 (researchers/reporting flag the malicious commit; attacker also contacted press)date_disclosed: 2025-07-23/24 (404 Media report; AWS bulletin AWS-2025-015 and GitHub advisory GHSA-7g7f-ff96-5gcw; clean v1.85.0 shipped)status: correctedagent_description: Amazon Q Developer extension for VS Code - agentic coding assistant with shell and AWS CLI execution on the developer's host; distributed by AWS through the VS Code marketplace at ~1M installs (marketplace count ~950K+).operator_type: enterprise (AWS-built and -distributed; end operators are individual developers)autonomy_level: human-on-the-loop typical (agent executes commands with configurable confirmation); the payload targeted whatever autonomy each install allowedmodel_stack: Amazon Q Developer hosted models (payload was model-agnostic natural language)harness: VS Code extension + open-source aws-toolkit-vscode repo + AWS CodeBuild release pipeline (the compromised link)authority_scope: code execution (shell on host), credentials (local AWS credentials), data access (home directory), cloud resources (EC2/S3/IAM via AWS CLI)funds_at_risk_usd: unknown/unbounded (fleet-wide destructive potential across ~1M installs; no meaningful cap existed)blast_radius: customers/third parties (every installed developer machine + their AWS accounts)root_cause: supply-chain-compromise (primary; v0.1 taxonomy - the compromised release pipeline is the failure); contributing prompt-injection (the payload's form: a natural-language instruction to the agent) and operator-error (overscoped GitHub token in the CodeBuild configuration granting a stranger admin-equivalent access)failure_locus: dependency (marketplace-distributed tool poisoned upstream of every operator)exploitation_status: in-wild-payload-failed (malicious artifact reached ~1M real installs for ~5-6 days; payload did not execute)mechanism: An outsider submitted a PR to aws-toolkit-vscode on ~July 13 and, via an inappropriately scoped GitHub token in the CodeBuild configuration, obtained commit access ("admin credentials on a silver platter," per the attacker to 404 Media; the access level is the attacker's and secondary reporting's description, not AWS's). They committed a prompt instructing the agent to "clean a system to a near-factory state" - wipe the home directory and delete EC2/S3/IAM resources via AWS CLI. The commit auto-built into official v1.84.0, published July 17. A syntax/formatting error prevented the injected prompt from executing. AWS revoked credentials, removed the code, and shipped v1.85.0.adversary_present: yes (the attacker told 404 Media the payload was deliberately defanged as a demonstration; that claim is the attacker's and is unverifiable)severity: near-miss (full distribution of a destructive payload; zero realized execution)direct_loss_usd: 0 reported by AWS (its bulletin states no changes to any services or customer environments; not an independently established zero)indirect_loss_usd: unknown (AWS incident response; customer update burden)downtime: none (v1.84.0 pulled; users required to update to 1.85.0)data_exposure: none reporteddetected_by: third-party / attacker-disclosure (security researchers flagged the commit; the attacker simultaneously self-disclosed to 404 Media) - not by AWS pipeline controlstime_to_detect: ~6-10 days from malicious commit; ~5-6 days of poisoned release livetime_to_recover: ~1-2 days from detection (credentials revoked, build pulled, v1.85.0 released)remediation: credential revocation, malicious code removal, forced version deprecation, security bulletinstructural_fix: release-pipeline credential scoping tightened (per AWS); the deeper fact stands - a marketplace agent's prompt channel is part of its supply chaincontrols_that_worked: none identified in AWS's bulletin; on the secondary reporting, the outcome was bounded by a syntax error in the payload (or, per the attacker, deliberate defanging), not by an AWS or user control. The "luck, not defense" reading is the registry's interpretation, recorded per v0.1 exploitation_status to avoid conflating it with defended near-misses.telemetry_grade: operator-logs (AWS's account), corroborated by public git history and the marketplace release recordsources:independence: good - first-party bulletins plus unaffiliated security press; attacker's account via 404 Media (referenced in the above).aiid_incident_id: 1158 (https://incidentdatabase.ai/cite/1158/) - cross-reference; primaries verified independentlyconfidence: high on timeline and mechanism; medium on "zero damage" (rests on AWS's attestation plus the attacker's claim - named weakest link)supply-chain-compromise with failure_locus dependency - the v0.1 amendment names this exact cluster. Prompt-injection kept as contributing (payload form).aiid_incident_id cross-reference (AIID 1158), matched against the AIID weekly database export (2026-08-17). A cross-reference, not a re-verification; no claim changed.direct_loss_usd reads as AWS's report rather than an established zero; controls_that_worked marks the luck reading as interpretation.See also - this event in the AI Incident Database: incident 1158.