PraisonAI is a multi-agent teams system. In versions prior to 4.6.40, the bundled Claude GitHub Actions workflow is vulnerable to command injection because it embeds an attacker-controlled pull request branch name into a Bash run: block without quoting or validation. Additionally, the workflow allows any @claude comment to trigger the job regardless of whether the commenter is a trusted collaborator. An outside contributor can open a pull request from a fork whose branch name contains shell metacharacters and comment @claude, causing Bash to execute arbitrary shell code in the GitHub Actions runner. Because these commands run in a job holding a GitHub App token with write permissions, OIDC access, and gh/git access, the injection can be chained through $GITHUB_PATH to compromise later privileged steps, enabling repository writes, pull request and issue manipulation, or OIDC-token abuse. This issue has been fixed in version 4.6.40.

GitPython before 3.1.52 is vulnerable to environment-variable exfiltration in Repo.clone_from(). The caller-supplied remote URL is passed through Git.polish_url(), which on non-Cygwin platforms calls os.path.expandvars() on the URL before invoking git clone. An attacker who controls the clone URL can embed $NAME or ${NAME} tokens that are expanded to the values of the hosting process's environment variables (e.g., AWS_SECRET_ACCESS_KEY or GITHUB_TOKEN). The resulting URL, now containing the secret, is transmitted over the network to an attacker-controlled host during the clone attempt, disclosing the secret.

Wazuh workflows before 44bf114 contain a shell injection vulnerability in GitHub Actions that allows attackers to execute arbitrary commands by submitting pull requests with crafted VERSION.json files. Attackers can inject shell metacharacters into environment variables that are directly interpolated into run steps, enabling command execution and exfiltration of secrets including GITHUB_TOKEN and AWS credentials on self-hosted runners.

GitHub MCP Server is GitHub's official MCP Server. Prior to 1.1.0, the CompletionsHandler function in pkg/github/server.go accesses params.Ref without first checking whether it is nil, so a completion/complete request with a missing or empty ref field triggers a nil pointer dereference and a Go runtime panic; because the crash occurs before any authentication or token validation, any unauthenticated client able to send JSON-RPC messages can crash the server, resulting in a complete denial of service. This issue is fixed in version 1.1.0.

cal.com (calcom repository, later renamed cal.diy) is affected by a repository takeover vulnerability in its GitHub Actions workflows. The workflow pr.yml uses the pull_request_target trigger with the repository's default write permissions and passes them down to check-types.yml. check-types.yml then performs a 'dangerous' checkout of the attacker-submitted pull request code (via the dangerous-git-checkout action) and subsequently executes it (through yarn install and package.json scripts). An attacker can open a pull request whose code runs arbitrary commands with the repository's write-scoped GITHUB_TOKEN, allowing them to push commits, merge or mutate pull requests, add or delete comments, and delete or force-push branches, thereby compromising the repository. The main branch is affected; no patched version is available.

Meshtastic is an open source mesh networking solution. Prior to version 2.7.21.1370b23, the Meshtastic GitHub repository's main_matrix.yml workflow is triggered by pull_request_target and multiple jobs check out the attacker's fork code and execute it with access to repository secrets and elevated GITHUB_TOKEN permissions. No approval gate exists. Pull requests from external users with author_association: "NONE" triggered the CI workflow automatically. The workflow directly executes attacker-controlled files from the fork checkout. This issue could have resulted in supply chain compromise, self-hosted runner compromise, and/or repository takeover for the repo. This issue is separate from GHSA-6mwm-v2vv-pp96, which addressed a command injection via github.head_ref in the setup job of the same workflow. That fix correctly moved to environment variables. However, the more critical fork checkout vulnerability across the check, build, and build-debian-src jobs was not addressed. Version 2.7.21.1370b23 contains a patch for thie issue.

In the Linux kernel, the following vulnerability has been resolved: Bluetooth: hci_sync: fix UAF in hci_le_create_cis_sync hci_le_create_cis_sync() dereferences conn->conn_timeout after releasing both rcu_read_lock() and hci_dev_lock(hdev). The conn pointer was obtained from an RCU-protected iteration over hdev->conn_hash.list and is not valid once these locks are dropped. A concurrent disconnect can free the hci_conn between the unlock and the dereference, causing a use-after-free read. The cancellation mechanism in hci_conn_del() cannot prevent this because hci_le_create_cis_pending() queues hci_create_cis_sync with data=NULL: hci_cmd_sync_queue(hdev, hci_create_cis_sync, NULL, NULL); While hci_conn_del() dequeues with data=conn: hci_cmd_sync_dequeue(hdev, NULL, conn, NULL); Since NULL != conn, the lookup in _hci_cmd_sync_lookup_entry() never matches, and the pending work item is not cancelled. Fix this by saving conn->conn_timeout into a local variable while the locks are still held, so the stale conn pointer is never dereferenced after unlock. This is the same class of bug as the one fixed by commit 035c25007c9e ("Bluetooth: hci_sync: Fix UAF on le_read_features_complete") which addressed the identical pattern in a different function. This vulnerability was identified using 0sec.ai, an open-source automated security auditing platform (https://github.com/0sec-labs).

In the Linux kernel, the following vulnerability has been resolved: amd-pstate: Fix memory leak in amd_pstate_epp_cpu_init() On failure to set the epp, the function amd_pstate_epp_cpu_init() returns with an error code without freeing the cpudata object that was allocated at the beginning of the function. Ensure that the cpudata object is freed before returning from the function. This memory leak was discovered by Claude Opus 4.6 with the aid of Chris Mason's AI review-prompts (https://github.com/masoncl/review-prompts/tree/main/kernel).

linux

KanaDojo contains a command injection vulnerability that allows an attacker with pull request access to execute arbitrary shell commands by inserting shell metacharacters into the version or changes fields of patchNotesData.json, which are interpolated unsanitized into a child_process.execSync() call in the release.yml workflow. Attackers can have a malicious pull request merged to trigger the GitHub Actions runner with contents write permissions and access to GITHUB_TOKEN.

KanaDojo before 0.1.18 contains a sandbox escape vulnerability that allows an attacker to execute arbitrary code by exploiting the explicit passing of the global require function into a Node.js vm.runInNewContext() sandbox context in the issue-auto-respond.yml workflow. Attackers can submit a pull request modifying messages.cjs to import arbitrary Node.js modules, bypassing sandbox restrictions and achieving remote code execution with full GitHub Actions runner privileges including access to AUTOMATION_PR_TOKEN.

A vulnerability was determined in TwiN gatus 5.36.0. Impacted is the function setSessionCookie of the file security/oidc.go of the component OIDC Session Cookie Handler. Executing a manipulation can lead to sensitive cookie without secure attribute. The attack can be launched remotely. This attack is characterized by high complexity. The exploitability is considered difficult. The reported GitHub issue was closed with the label "not planned".

Rejected reason: Retracted following review by Red Hat Product Security and confirmation from the upstream Clair/Claircore maintainer. This CVE misattributes the described behavior to github.com/quay/claircore: the authentication mechanism in question (optional PSK, HTTP endpoint /indexer/api/v1/index_report) is implemented entirely in github.com/quay/clair; no PSK-related code exists anywhere in claircore's codebase or git history. The unauthenticated indexer API is Clair's documented, intentional design, authentication is an opt-in deployment choice, not a code defect. No fix commit was found in claircore between the version recorded as the affected boundary (1.5.52) and the following release (1.5.53); intervening commits are unrelated dependency and feature changes, so the "fixed in 1.5.52" status is inaccurate.

A vulnerability was found in macrozheng mall up to 1.0.3. This affects an unknown function of the file /admin/update/ of the component Super Admin Password Handler. Performing a manipulation results in improper authorization. Remote exploitation of the attack is possible. The vendor deleted the GitHub issue for this vulnerability without any explanation. Afterwards the vendor was contacted early about this disclosure via email but did not respond in any way.

GitHub CLI (gh) is GitHub’s official command line tool. Prior to 2.93.0, GitHub CLI incorrectly includes authorization header in API requests to TUF repository mirrors via gh attestation, gh release verify, and gh release verify-asset commands. The CLI uses a shared HTTP client with an authentication layer that automatically attaches tokens to outgoing requests. This layer lacks accurate host detection and can incorrectly attribute the target host, providing it with a token it should never receive. Specifically, the host normalization logic collapses any *.github.com subdomain to github.com, so a request to tuf-repo.github.com (a GitHub Pages site, not a GitHub API endpoint) is treated as a request to github.com and receives the user's github.com token. For hosts that don't match github.com or a known GHES instance at all, the resolver falls back to GH_ENTERPRISE_TOKEN if set. The gh attestation, gh release verify and gh release verify-asset commands fetch data from several external hosts as part of their normal operation (TUF metadata from tuf-repo.github.com and tuf-repo-cdn.sigstore.dev, artifact bundles from Azure Blob Storage). Because these requests go through the same authenticated HTTP client, the token is sent to all of them. This vulnerability is fixed in 2.93.0.

Espressif Shared GitHub DangerJS is a reusable GitHub Action CI DangerJS workflow for Espressif GitHub projects. Prior to 1.0.1, the action's entrypoint.sh invoked DangerJS from the caller's workspace after copying the fork's checkout into it, creating an untrusted search path for both binary resolution and Node.js module resolution. A fork pull request processed by a pull_request_target workflow could therefore cause fork-supplied code to execute inside the action container in place of the action's own code. This vulnerability is fixed in 1.0.1.

Sherlock hunts down social media accounts by username across social networks. Prior to 0.16.1, the GitHub Actions workflow validate_modified_targets.yml is vulnerable to command injection via the pull_request_target trigger. Any GitHub user can execute arbitrary commands on the CI runner and exfiltrate the GITHUB_TOKEN by opening a pull request. No approval, review, or merge is required. This vulnerability is fixed in 0.16.1.

A stored cross-site scripting (XSS) vulnerability in the /admin/config-module.php component of creatorsofcode simplephp GitHub commit 5184cff (Latest as of 2026-02-27) via injecting a crafted payload.

A cross-site request forgery (CSRF) vulnerability in Jenkins GitHub Integration Plugin 0.7.3 and earlier allows attackers to attackers to trigger a build for a pull request.

kostyasha

GuardDog is a CLI tool to identify malicious PyPI packages. From 1.0.0 to 2.9.0, the programmatic remote project scanning path rewrites attacker-controlled repository URLs using a blind string replacement and then sends the caller's GitHub credentials with the resulting request. This allows an attacker who can influence the scanned repository URL to trigger SSRF and capture the GH_TOKEN used by GuardDog. This vulnerability is fixed in .

In the Linux kernel, the following vulnerability has been resolved: xfrm: fix ip_rt_bug race in icmp_route_lookup reverse path icmp_route_lookup() performs multiple route lookups to find a suitable route for sending ICMP error messages, with special handling for XFRM (IPsec) policies. The lookup sequence is: 1. First, lookup output route for ICMP reply (dst = original src) 2. Pass through xfrm_lookup() for policy check 3. If blocked (-EPERM) or dst is not local, enter "reverse path" 4. In reverse path, call xfrm_decode_session_reverse() to get fl4_dec which reverses the original packet's flow (saddr<->daddr swapped) 5. If fl4_dec.saddr is local (we are the original destination), use __ip_route_output_key() for output route lookup 6. If fl4_dec.saddr is NOT local (we are a forwarding node), use ip_route_input() to simulate the reverse packet's input path 7. Finally, pass rt2 through xfrm_lookup() with XFRM_LOOKUP_ICMP flag The bug occurs in step 6: ip_route_input() is called with fl4_dec.daddr (original packet's source) as destination. If this address becomes local between the initial check and ip_route_input() call (e.g., due to concurrent "ip addr add"), ip_route_input() returns a LOCAL route with dst.output set to ip_rt_bug. This route is then used for ICMP output, causing dst_output() to call ip_rt_bug(), triggering a WARN_ON: ------------[ cut here ]------------ WARNING: net/ipv4/route.c:1275 at ip_rt_bug+0x21/0x30, CPU#1 Call Trace: <TASK> ip_push_pending_frames+0x202/0x240 icmp_push_reply+0x30d/0x430 __icmp_send+0x1149/0x24f0 ip_options_compile+0xa2/0xd0 ip_rcv_finish_core+0x829/0x1950 ip_rcv+0x2d7/0x420 __netif_receive_skb_one_core+0x185/0x1f0 netif_receive_skb+0x90/0x450 tun_get_user+0x3413/0x3fb0 tun_chr_write_iter+0xe4/0x220 ... Fix this by checking rt2->rt_type after ip_route_input(). If it's RTN_LOCAL, the route cannot be used for output, so treat it as an error. The reproducer requires kernel modification to widen the race window, making it unsuitable as a selftest. It is available at: https://gist.github.com/mrpre/eae853b72ac6a750f5d45d64ddac1e81

In the Linux kernel, the following vulnerability has been resolved: serial: caif: fix use-after-free in caif_serial ldisc_close() There is a use-after-free bug in caif_serial where handle_tx() may access ser->tty after the tty has been freed. The race condition occurs between ldisc_close() and packet transmission: CPU 0 (close) CPU 1 (xmit) ------------- ------------ ldisc_close() tty_kref_put(ser->tty) [tty may be freed here] <-- race window --> caif_xmit() handle_tx() tty = ser->tty // dangling ptr tty->ops->write() // UAF! schedule_work() ser_release() unregister_netdevice() The root cause is that tty_kref_put() is called in ldisc_close() while the network device is still active and can receive packets. Since ser and tty have a 1:1 binding relationship with consistent lifecycles (ser is allocated in ldisc_open and freed in ser_release via unregister_netdevice, and each ser binds exactly one tty), we can safely defer the tty reference release to ser_release() where the network device is unregistered. Fix this by moving tty_kref_put() from ldisc_close() to ser_release(), after unregister_netdevice(). This ensures the tty reference is held as long as the network device exists, preventing the UAF. Note: We save ser->tty before unregister_netdevice() because ser is embedded in netdev's private data and will be freed along with netdev (needs_free_netdev = true). How to reproduce: Add mdelay(500) at the beginning of ldisc_close() to widen the race window, then run the reproducer program [1]. Note: There is a separate deadloop issue in handle_tx() when using PORT_UNKNOWN serial ports (e.g., /dev/ttyS3 in QEMU without proper serial backend). This deadloop exists even without this patch, and is likely caused by inconsistency between uart_write_room() and uart_write() in serial core. It has been addressed in a separate patch [2]. KASAN report: ================================================================== BUG: KASAN: slab-use-after-free in handle_tx+0x5d1/0x620 Read of size 1 at addr ffff8881131e1490 by task caif_uaf_trigge/9929 Call Trace: <TASK> dump_stack_lvl+0x10e/0x1f0 print_report+0xd0/0x630 kasan_report+0xe4/0x120 handle_tx+0x5d1/0x620 dev_hard_start_xmit+0x9d/0x6c0 __dev_queue_xmit+0x6e2/0x4410 packet_xmit+0x243/0x360 packet_sendmsg+0x26cf/0x5500 __sys_sendto+0x4a3/0x520 __x64_sys_sendto+0xe0/0x1c0 do_syscall_64+0xc9/0xf80 entry_SYSCALL_64_after_hwframe+0x77/0x7f RIP: 0033:0x7f615df2c0d7 Allocated by task 9930: Freed by task 64: Last potentially related work creation: The buggy address belongs to the object at ffff8881131e1000 which belongs to the cache kmalloc-cg-2k of size 2048 The buggy address is located 1168 bytes inside of freed 2048-byte region [ffff8881131e1000, ffff8881131e1800) The buggy address belongs to the physical page: page_owner tracks the page as allocated page last free pid 9778 tgid 9778 stack trace: Memory state around the buggy address: ffff8881131e1380: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb ffff8881131e1400: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb >ffff8881131e1480: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb ^ ffff8881131e1500: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb ffff8881131e1580: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb ================================================================== [1]: https://gist.github.com/mrpre/f683f244544f7b11e7fa87df9e6c2eeb [2]: https://lore.kernel.org/linux-serial/20260204074327.226165-1-jiayuan.chen@linux.dev/T/#u

The Github Shortcode plugin for WordPress is vulnerable to Stored Cross-Site Scripting via the 'repo' shortcode attribute in the 'github' shortcode in all versions up to, and including, 0.1 due to insufficient input sanitization and output escaping. This makes it possible for authenticated attackers, with Contributor-level access and above, to inject arbitrary web scripts in pages that will execute whenever a user accesses an injected page.