AI agents invoke runPipeline to trigger actions in Bitbucket MCP. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call — builds kicked off, notifications sent, workflows started.
Pipelines in Bitbucket execute build, test, and deployment workflows defined in repository configuration. Triggering a pipeline executes code with side effects that depend on pipeline definition (deploy scripts, external API calls, data mutations). This is Execute rather than Write because the effects are determined by code execution rather than direct data modification.
From the tool's definition Tool explicitly 'Trigger[s] a new pipeline run' which executes CI/CD workflows on arbitrary code, potentially deploying to production, running tests, or triggering external integrations.
Documented attack patterns abuse exactly the kind of access runPipeline gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Bitbucket MCP, and nothing reaches the server without passing your rules. This is the rule we recommend for runPipeline:
{
"version": "1",
"default": "deny",
"tools": {
"runPipeline": {
"limits": [
{
"counter": "runpipeline_rate",
"window": "minute",
"max": 10,
"scope": "grant"
}
]
}
}
}runPipeline stays usable, but rate-capped — a runaway agent can't fire it dozens of times a minute. Everything else on the server is denied unless you say otherwise.
Free to start. No card required.
Trigger a new pipeline run. It is categorised as a Execute tool in the Bitbucket MCP MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Bitbucket MCP server in PolicyLayer and add a rule for runPipeline: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches Bitbucket MCP. Nothing to install.
runPipeline is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.
Yes. Add a rate_limit block to the runPipeline rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for runPipeline. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
runPipeline is provided by the Bitbucket MCP server (matanyemini/bitbucket-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Bitbucket MCP, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.
Free to start. No card required.
49 Bitbucket MCP tools catalogued and risk-classified — across an index of 43,000+ MCP servers.