Skip to content

    Type a page, a command such as keryn scan, or a rule id.

    recursive-agent-loop-safety-v2

    View as Markdown

    6 rules · Keryn Software Licence · rules 0.8.0 · all packs

    How to read a rule
    Severity
    How bad the finding is if it is real: error fails a default gate, warning and info do not.
    Confidence
    P0 exploitable as written; P1 dangerous in context; P2 a heuristic or a posture check: confirm by reading the code.
    False-positive risk
    low: an exact API or flag, so hits are almost always real; medium: the pattern is legitimate in many programs, so check each hit; high: it finds a lead to confirm, so expect benign hits.
    Edge
    The relationship the finding sits on. reaches: content or traffic gets to a thing; calls: one thing invokes another; reads: one thing reads from another; writes: one thing writes to another; sends: data leaves for another party; grants: authority is given or left unchecked; depends_on: something relies on a component it does not verify; trusts: one party believes another without evidence. The boundary it crosses, when there is one: user->prompt (untrusted input reaching a prompt); model->code (model output reaching code, a shell, a query or a page); secret->outside (a secret leaving the trusted side); supply-chain (code, models or prompts pulled in from outside).
    Taxonomy chips
    References for readers who work in OWASP LLM 2025, OWASP Agentic, CWE or NIST AI RMF. They say where a rule belongs, not that a framework is satisfied.

    seceng.loop.agent-loop-without-max-steps-ts

    Section titled “seceng.loop.agent-loop-without-max-steps-ts”

    warning typescript, javascript recursive-agent-loop-safety

    Agent/tool loop appears to run without visible maxSteps, maxIterations, depth, timeout, cancellation, or budget guard.

    Confidence P2: a heuristic or a posture check: confirm by reading the code. False-positive risk medium: the pattern is legitimate in many programs, so check each hit.

    Edge inspected: grants authority is given or left unchecked.

    CWE-770 LLM06 LLM10 ASI02 ASI08 MANAGE

    Also reported under the older id seceng.llm10.agent-loop-without-obvious-limit-js.

    seceng.loop.agent-loop-without-max-steps-python

    Section titled “seceng.loop.agent-loop-without-max-steps-python”

    warning python recursive-agent-loop-safety

    Agent/tool loop appears to run without visible max_steps, max_iterations, depth, timeout, cancellation, or budget guard.

    Confidence P2: a heuristic or a posture check: confirm by reading the code. False-positive risk medium: the pattern is legitimate in many programs, so check each hit.

    Edge inspected: grants authority is given or left unchecked.

    CWE-770 LLM06 LLM10 ASI02 ASI08 MANAGE

    Also reported under the older id seceng.llm10.agent-loop-without-obvious-limit-py.

    seceng.loop.langchain-agent-no-max-iterations

    Section titled “seceng.loop.langchain-agent-no-max-iterations”

    warning python, typescript, javascript langchain

    LangChain agent executor is configured without max_iterations, timeout, or early stopping controls.

    Confidence P2: a heuristic or a posture check: confirm by reading the code. False-positive risk medium: the pattern is legitimate in many programs, so check each hit.

    Edge inspected: grants authority is given or left unchecked.

    CWE-770 LLM06 LLM10 ASI02 ASI08 MANAGE

    seceng.loop.langgraph-recursion-limit-missing

    Section titled “seceng.loop.langgraph-recursion-limit-missing”

    warning python, typescript, javascript langgraph

    LangGraph invocation appears to omit recursion_limit or equivalent loop guard.

    Confidence P2: a heuristic or a posture check: confirm by reading the code. False-positive risk medium: the pattern is legitimate in many programs, so check each hit.

    Edge inspected: grants authority is given or left unchecked.

    CWE-770 LLM06 LLM10 ASI08 ASI02 MANAGE

    seceng.loop.recursive-tool-calls-model-again

    Section titled “seceng.loop.recursive-tool-calls-model-again”

    info typescript, javascript, python recursive-agent-loop-safety

    Tool handler appears to call the model again, enabling recursive tool/model loops without explicit guard.

    Confidence P2: a heuristic or a posture check: confirm by reading the code. False-positive risk medium: the pattern is legitimate in many programs, so check each hit.

    Edge inspected: calls one thing invokes another.

    CWE-770 LLM06 LLM10 ASI02 ASI08 MANAGE

    seceng.loop.streaming-without-cancel-or-timeout

    Section titled “seceng.loop.streaming-without-cancel-or-timeout”

    info typescript, javascript, python, java, csharp recursive-agent-loop-safety

    Streaming LLM response appears to run without visible cancellation, timeout, or abort signal.

    Confidence P1: dangerous in context. False-positive risk high: it finds a lead to confirm, so expect benign hits.

    Edge inspected: grants authority is given or left unchecked.

    CWE-770 LLM10 ASI08 MANAGE