Back to skill

Security audit

Forgejo Workflow

Security checks for vulnerabilities and agentic risk

Overview

This Forgejo skill is a disclosed workflow helper for authenticated repository operations, with expected but sensitive write/delete capabilities users should handle carefully.

Install only for Forgejo instances where you are comfortable giving the agent repository access. Use the narrowest token scopes needed, avoid pasting or logging tokens, and confirm destructive actions such as repository deletes, wiki deletes, label or milestone deletes, merges, and workflow dispatches before allowing them.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (9)

Credential Access

High
Category
Privilege Escalation
Confidence
60% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · README.md (reported line 73)May include surrounding context.

If you already have credentials:

text
Set up this Forgejo skill for https://<your-forgejo-host>. I already have a token and my Forgejo username is alice. First check whether fj, curl, jq, and git are available. If one of them is missing and OpenClaw can install it, install it. Then assume FORGEJO_URL and FORGEJO_TOKEN may already be available either from the OpenClaw .env loaded for this session or from exported shell variables, verify whether they are present, tell me the exact fj --host <your-forgejo-host> auth add-key alice command to run if fj auth is not ready yet, and finally tell me clearly whether the skill is fully ready, partially ready, or blocked, and what is still missing if anything.

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
80% 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 · references/api-cheatsheet.md (reported line 60)May include surrounding context.

Delete a page:

http
DELETE /repos/{owner}/{repo}/wiki/page/{pageName}

List or read pages:

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
80% 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 · references/api-cheatsheet.md (reported line 115)May include surrounding context.

Delete a label:

http
DELETE /repos/{owner}/{repo}/labels/{id}

Milestones

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
80% 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 · references/api-cheatsheet.md (reported line 151)May include surrounding context.

Delete a milestone:

http
DELETE /repos/{owner}/{repo}/milestones/{id}

Pull request reviews with inline comments

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The README instructs users to export and use a Forgejo application token but does not warn against exposing it in shell history, logs, screenshots, shared terminals, or committing it into files. Because this skill is specifically designed to operate with authenticated Forgejo access, mishandling the token could directly enable repository, issue, PR, and CI access against the target instance.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

This markdown file documents DELETE operations for wiki pages, labels, and later milestones, but it does not include any warning that these actions can remove repository data or metadata. Under the markdown-specific missing-warning rule, user-facing documentation should warn when behaviors may affect user data or system integrity.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · references/api-cheatsheet.md (reported line 255)May include surrounding context.

bash
page=1
while true; do
  curl -sS -D headers.txt \
    -H "Authorization: token $FORGEJO_TOKEN" \
    -H "Accept: application/json" \
    "$FORGEJO_URL/api/v1/repos/$OWNER/$REPO/issues?page=$page&limit=50"

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
75% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · references/code-review.md (reported line 187)May include surrounding context.

md
## Common pitfalls

- Reviewing style instead of behavior
- Assuming the base branch is `main` without checking PR metadata
- Missing removed-line comments that require `old_position` instead of `new_position`
- Approving by default instead of keeping approval separate from comments
- Overstating certainty when the evidence is incomplete

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
79% confidence
Finding

The file instructs use of FORGEJO_TOKEN in Authorization headers but does not warn readers to protect the token or avoid exposing it in logs, shell history, or shared outputs. Because the markdown describes behavior involving credentials and authenticated API calls, a short disclosure about sensitive token handling is expected.

Content

No source excerpt is available for this finding.