Finish an authorized security research task with a durable solved, abstained, or blocked outcome. Use when evidence is sufficient, a stop condition is reached, a tool budget is nearly exhausted, or the agent must commit its best supported answer.
中文简介正在准备中。当前显示原文说明。
查看安装方法安装前可以查看交给 AI 代理的指令内容。
Finish deliberately. A provider stopping or a prose summary is not a committed research outcome.
security-artifact-manifest / security-artifact-access and collect the resulting artifact IDs. Verify every ID: if a tool returns no artifactId, persistence failed — retry or finish blocked; never proceed as if the evidence is saved. Check persistedContentSource: an artifact persisted from manifest text alone (no workspace file content) is a summary stub, not the evidence bytes — treat it as unsaved and do not cite it.update_research_state with the final branch dispositions, artifact-backed evidence summaries, best candidate answer, and any remaining blocker. This is an upsert: include changed entries, not a transcript rewrite. Never invent an artifact ID.solved: a concrete answer — including a well-supported negative result such as "X is not vulnerable to Y" — is supported by durable evidence;abstained: available evidence does not support an answer either way; state the null answer, what was attempted, and the limiting evidence gap;blocked: authorization, approval, target, tool, or runtime state prevents completion.Call finish_research exactly once with the best supported outcome. Solved outcomes need a concrete answer and evidence artifact IDs. Blocked outcomes need a concise remaining-uncertainty explanation. Confidence is a calibrated value from 0 to 1, not a benchmark score.
If the tool rejects the call, repair only the named fields or blocker and retry within the reported limit. Do not evade the gate in prose or begin unrelated exploration. If repairs are exhausted, stop as blocked.
A prose conclusion, empty response, or thought-only stop is not completion. The controller records it as a malformed terminal attempt and may issue one bounded recovery/wrap-up turn. Respond with the typed tool call; do not repeat the prose summary.
Hidden benchmark answers and scorer feedback are never research inputs. The terminal call records what the research agent committed; a separate product review or deterministic oracle decides whether it succeeded.
At a wrap-up checkpoint, choose among submitting the best supported result, abstaining, or performing one decisive remaining check. Before that check, upsert update_research_state with the current plan of record so an interruption leaves a durable resumable pause rather than a lost plan. Preserve open uncertainty and evidence provenance. Never spend the final budget on a new broad branch.
还没有评价。欢迎分享你的使用感受。
无需登录即可发表。欢迎分享使用感受、提问，或说明在哪些环境下无法运行。
还没有评论。来写第一条吧。
Plan and use user-consented local camera and microphone observations for an explicitly authorized physical target. Use when a capture can establish a bounded physical-state observation such as whether a test fixture changed state, an instrument reading changed, or equipment noise changed relatively.
中文简介正在准备中。当前显示原文说明。
Plan vulnerability research using conventional bug-hunting and pentest stage names, specialty playbooks, and a research-hypothesis-experiment loop. Use when designing a bug bounty workflow, selecting tools and strategies, building a decision tree, or turning a target/scope into a staged hunting plan.
中文简介正在准备中。当前显示原文说明。
Passive code-review lane for authorized repositories, local checkouts, uploaded source archives, and diffs. Use when the security research agent needs deterministic candidate discovery before expensive AI investigation, scoped PR-style review, append-only evidence records, or independent revalidation of code findings.
中文简介正在准备中。当前显示原文说明。
Run independent, task-matched expert attempts, select a base, and synthesize the strongest result.
中文简介正在准备中。当前显示原文说明。
Select established CTF and security-research tools for authorized artifact triage, web inspection, reversing, crypto, forensics, networking, and exploit validation. Use when a user provides a CTF-style tool catalog, asks which tools to use, or needs safe usage instructions for common offensive/security tools.
中文简介正在准备中。当前显示原文说明。
Disciplined diagnosis loop for hard bugs and performance regressions. Reproduce → minimise → hypothesise → instrument → fix → regression-test. Use when user says "diagnose this" / "debug this", reports a bug, says something is broken/throwing/failing, or describes a performance regression.
中文简介正在准备中。当前显示原文说明。