Back to skill

Security audit

Lingxi · Daily Ops Automation

Security checks for vulnerabilities and agentic risk

Overview

This skill is not clearly malicious, but it asks for broad GitHub account access and stores the token in a local config while automating public actions, so it should be reviewed carefully.

Install only if you are comfortable with automated actions being taken under your GitHub identity. Use a fine-grained, least-privilege token or GitHub App credential, avoid plaintext token storage where possible, restrict any config file permissions, review all target repositories and publishing channels, and prefer a dry-run or manual approval step before enabling scheduled runs.

Vulnerability Patterns
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • 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
Findings (1)

T09 · Insecure Skill Coding Practices

Error
Location
SKILL.md:62
Finding

Plaintext Storage of an Overprivileged GitHub Access Token

Content
View full analysis

Vulnerability Details

File Location: SKILL.md, lines 62–68 and 133–137
Vulnerability Type: Plaintext credential storage and excessive token permissions
Risk Level: High

Vulnerable Code

SKILL.md, lines 62–68:

markdown
Create `~/.openclaw/daily-ops-config.json`:

```json
{
  "github_token": "ghp_xxx",
  "watch_topics": ["ai", "open-source", "developer-tools"],

SKILL.md, lines 133–137:

markdown
## Requirements

- `gh` CLI authenticated with appropriate scopes
- GitHub token with: `repo`, `read:user`, `public_repo`
- OpenClaw with `github` skill installed

Technical Analysis

The Skill instructs users to store a GitHub access token directly in a plaintext JSON configuration file under the user's home directory. It does not require restrictive file permissions, encryption, an operating-system credential store, or a dedicated secrets manager. Consequently, the token may be exposed to other local users, processes, backup systems, diagnostic tools, or accidental configuration-file disclosure.

The requested permissions also exceed the minimum privileges evidently required by the declared workflow. The broad classic repo scope can grant access to private repositories and repository write operations, while the described activities primarily concern public trending projects, stars, comments, issue updates, and report publication. The simultaneous request for repo and public_repo is redundant because repo encompasses public-repository access.

No real token is embedded in the package—the displayed ghp_xxx value is a placeholder. The vulnerability arises from instructing users to persist their actual credential insecurely and to grant it unnecessarily broad authority.

Attack Path

  1. A user configures the Skill by placing a real GitHub token in ~/.openclaw/daily-ops-config.json.
  2. The user grants the documented repo, read:user, and public_repo scopes.

...[truncated 1433 chars]

Remediation
View remediation

Remediation Suggestions

  1. Do not store GitHub tokens directly in daily-ops-config.json. Keep only a secret reference or credential identifier in the configuration file.
  2. Retrieve credentials at runtime from an operating-system keychain, a dedicated secrets manager, or GitHub CLI's protected authentication store.
  3. If file-based storage is unavoidable, place the secret in a separate file, require permission mode 0600, verify ownership before use, and prevent the file from being committed, logged, backed up insecurely, or included in reports.
  4. Prefer a fine-grained personal access token or GitHub App installation token restricted to explicitly configured repositories.
  5. Grant only the permissions needed for enabled operations. Do not request private-repository access when the workflow targets only public repositories.
  6. Remove the redundant combination of repo and public_repo; avoid the broad classic repo scope wherever possible.
  7. Separate read-only trending and health-check operations from write operations. Request write credentials only when starring, commenting, updating issues, or uploading reports is explicitly enabled.
  8. Add token expiration, rotation, and revocation guidance, and instruct users to review the GitHub audit log after suspected exposure.
  9. Redact credentials from generated reports, command output, exceptions, logs, and social-platform publishing content.
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Rogue AgentSelf-Modification, Session Persistence
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill description advertises automated GitHub interactions and multi-platform publishing but does not warn users that it will perform external actions and transmit data to third-party services. This can cause users to authorize or run the skill without understanding that it may post content, star repositories, and contact remote endpoints on their behalf.

Content

No source excerpt is available for this finding.

Session Persistence

Medium
Category
Rogue Agent
Confidence
88% confidence
Finding

The skill directs users to persist operational configuration, including a GitHub token, under ~/.openclaw/daily-ops-config.json, which implies long-lived local storage of credentials and automation settings. If the file is readable by other local users, backed up insecurely, or later accessed by other tools, the token could be stolen and used to perform GitHub actions as the user.

Content

Scanner excerpt · SKILL.md (reported line 64)May include surrounding context.

Configuration

Create ~/.openclaw/daily-ops-config.json:

json
{

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The configuration block instructs users to place a GitHub token in a local JSON file and defines external targets, but it provides no caution about secure storage, token scope minimization, or the risks of automated outbound actions. This increases the chance of credential leakage, over-privileged tokens, or unintended use of the user's GitHub identity for automated activity.

Content

No source excerpt is available for this finding.