Back to skill

Security audit

Todo

Security checks for vulnerabilities and agentic risk

Overview

ClawScan rates this Todo skill benign with medium confidence: the reported findings appear related to its stated task-management purpose, though one scanner was partial and some risks remain to review.

The verdict is benign, not a guarantee of safety. Consider whether you are comfortable with the skill’s command-trigger behavior and its unpinned `npx jest` examples. Coverage was partial: 10 of 11 components were fully inspected, and tool-misuse analysis was degraded, so the evidence is incomplete.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (7)

Shadow Command Trigger

Medium
Category
Trigger Abuse
Confidence
70% confidence
Finding

Shadow Command Trigger: description clause 'Invoked directly as /todo it only prints the command list and a read-only invariant check' conflicts with built-in command 'list'

Content

No source excerpt is available for this finding.

MCP server referenced without pinned version: 'npx jest'.

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding

npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Content

No source excerpt is available for this finding.

MCP server referenced without pinned version: 'npx jest'.

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding

npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Content

No source excerpt is available for this finding.

MCP server referenced without pinned version: 'npx jest'.

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding

npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Content

No source excerpt is available for this finding.

MCP server referenced without pinned version: 'npx jest'.

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding

npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Content

No source excerpt is available for this finding.

Tool Parameter Abuse

Low
Category
Tool Misuse
Confidence
15% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · SKILL.md (reported line 202)May include surrounding context.

md
| I7 | Every ID in `Depends:` or `From:` exists in TODO.md or CHANGELOG.md |
| I8 | Tree integrity: child ID = parent ID + `.n`, depth at most 3, even indent, no `[x]` root, no `[x]` node with open children, and any size given matches the tree |
| I9 | No dependency cycles, counting "parent waits for its children" |
| I10 | No forbidden command in `Done when`: `sudo`, `rm -r`/`-f`, `git push`, `git reset --hard`, `git clean`, `chmod -R`, `mkfs`, `dd if=`, the shell builtins `eval` and `exec`, a download whose output goes straight to a shell, one-liner code flags (`sh`/`bash`/`zsh`/`python` -c, `node`/`ruby`/`perl` -e, `php` -r), `find -exec rm` and `-delete`, publish, upload and deploy commands (`npm publish`, `twine upload`, `docker push`, `mvn deploy` and the like), and reads of credential files (SSH keys, cloud and tool credentials) |

**The invariant check** ("lint" in the skills) means: if `.todo/todo-lint.sh` exists in the project, run `bash .todo/todo-lint.sh` from the repositor

Tool Parameter Abuse

Low
Category
Tool Misuse
Confidence
15% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · SKILL.md (reported line 202)May include surrounding context.

md
| I7 | Every ID in `Depends:` or `From:` exists in TODO.md or CHANGELOG.md |
| I8 | Tree integrity: child ID = parent ID + `.n`, depth at most 3, even indent, no `[x]` root, no `[x]` node with open children, and any size given matches the tree |
| I9 | No dependency cycles, counting "parent waits for its children" |
| I10 | No forbidden command in `Done when`: `sudo`, `rm -r`/`-f`, `git push`, `git reset --hard`, `git clean`, `chmod -R`, `mkfs`, `dd if=`, the shell builtins `eval` and `exec`, a download whose output goes straight to a shell, one-liner code flags (`sh`/`bash`/`zsh`/`python` -c, `node`/`ruby`/`perl` -e, `php` -r), `find -exec rm` and `-delete`, publish, upload and deploy commands (`npm publish`, `twine upload`, `docker push`, `mvn deploy` and the like), and reads of credential files (SSH keys, cloud and tool credentials) |

**The invariant check** ("lint" in the skills) means: if `.todo/todo-lint.sh` exists in the project, run `bash .todo/todo-lint.sh` from the repositor