Security audit
Security checks for vulnerabilities and agentic risk
This DevRev skill is coherent, but it can change live DevRev records with a PAT and does not clearly require confirmation or safe token handling.
Install only if you want the agent to access DevRev on your behalf. Use a least-privilege PAT, store it as an environment variable rather than pasting it into chat, and require the agent to show the target record ID and request explicit approval before create, update, or delete operations.
Remote code is downloaded and executed. This bypasses code review and could introduce malicious code.
curl -s "https://api.devrev.ai/works.list" \
-H "Authorization: $DEVREV_TOKEN" | python3 -m json.tool
Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.
DevRev uses PAT (Personal Access Token) for authentication.
Header: Authorization: <token>
The skill description is broad enough to trigger on generic DevRev-related requests, including ones that may cause unintended remote reads or writes. Because the skill includes create and update operations against a live external service, overbroad invocation increases the chance of unauthorized or accidental actions.
The description says the skill can create and update DevRev issues and tickets but does not prominently warn that it performs state-changing operations on a remote system. That omission can mislead an agent or user into invoking the skill without understanding that it may modify production data.
The setup text instructs the user to provide a PAT token but gives no handling guidance for this sensitive credential. In agent environments, this increases the risk of secrets being pasted into chat, logged, echoed in shell history, or exposed through debugging output.
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
curl -s "https://api.devrev.ai/works.list" \
-H "Authorization: $DEVREV_TOKEN" | python3 -m json.tool
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
curl -s "https://api.devrev.ai/works.list" \
-H "Authorization: $DEVREV_TOKEN" | python3 -m json.tool
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
curl -s "https://api.devrev.ai/works.list" \
-H "Authorization: $DEVREV_TOKEN" | python3 -m json.tool
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
curl -s "https://api.devrev.ai/works.list" \
-H "Authorization: $DEVREV_TOKEN" | python3 -m json.tool
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
curl -s "https://api.devrev.ai/works.list" \
-H "Authorization: $DEVREV_TOKEN" | python3 -m json.tool
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
curl -s "https://api.devrev.ai/works.list" \
-H "Authorization: $DEVREV_TOKEN" | python3 -m json.tool
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
curl -s "https://api.devrev.ai/works.list" \
-H "Authorization: $DEVREV_TOKEN" | python3 -m json.tool
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
Filter by type:
# Only issues
curl -s "https://api.devrev.ai/works.list?type[]=issue" \
-H "Authorization: $DEVREV_TOKEN"
# Only tickets
This example performs a remote create operation against DevRev, which can change external system state and create records if triggered unintentionally. In the context of an agent skill, such mutating network actions are more dangerous than read-only calls because broad invocation or weak confirmation can lead to unauthorized or accidental issue creation.
curl -s -X POST "https://api.devrev.ai/works.create" \
-H "Authorization: $DEVREV_TOKEN" \
-H "Content-Type: application/json" \
-d '{
This API example creates a ticket in a remote system, so misuse can generate unauthorized records, workflow noise, or business process disruption. The context makes it moderately risky because the skill is designed for live operational changes but lacks strong safety framing around mutation.
curl -s -X POST "https://api.devrev.ai/works.create" \
-H "Authorization: $DEVREV_TOKEN" \
-H "Content-Type: application/json" \
-d '{
Updating an existing work item is a state-changing API action that can alter titles, bodies, and potentially operational workflow data. If invoked from ambiguous prompts or with the wrong identifier, it can damage or overwrite legitimate records in DevRev.
curl -s -X POST "https://api.devrev.ai/works.update" \
-H "Authorization: $DEVREV_TOKEN" \
-H "Content-Type: application/json" \
-d '{
This markdown file describes credential-based authentication and includes a token format example and shell usage, but it does not warn users to protect PATs or avoid leaking them in terminals, commits, or shared snippets. For markdown files, SQP-2 applies when descriptions omit warnings about behaviors that could affect privacy or system integrity.
No suspicious patterns detected.