菜单
Use when choosing short AgentCanon tool, skill, profile, check, runtime, closeout, or evidence routes from long candidate names, broad workflow text, routing misses, over-constrained related-skill candidates, public/system skill delegation, skill splitting, or skill/tool routing refactors.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
| name | task-routing |
| description | Use when choosing short AgentCanon tool, skill, profile, check, runtime, closeout, or evidence routes from long candidate names, broad workflow text, routing misses, over-constrained related-skill candidates, public/system skill delegation, skill splitting, or skill/tool routing refactors. |
Use the command packet before applying this skill's workflow:
python3 tools/agent_tools/skill_tool_commands.py show --skill task-routing --format text
Execute the required and task-matching conditional commands that the packet prints.
agents/skills/task-routing.md.python3 tools/agent_tools/route.py --name <candidate> to resolve a long proposed tool or skill name before creating any new public surface.python3 tools/agent_tools/route.py --prompt "<user request>" --format json when a broad request needs deterministic public skill routing. Treat ACTIVE_SKILLS as current-wave skill guidance, DEFERRED_SKILLS as selected later wave triggers, and RELATED_SKILL_CANDIDATES as evidence-gated candidates for later stages.$agent-log-analysis, durable issue candidates to $issue-finding-report, and recurrence feedback to $agent-learning.$openai-docs, $skill-creator, $skill-installer, $imagegen, or $plugin-creator; keep AgentCanon changes to local routing, evidence, and owner-surface contracts.python3 tools/agent_tools/route.py --area <area> --changed <paths...> to select the structured route for surface, profile, checks, environment, remote, AgentCanon update, MCP, goal, runtime, token, skill, agent, closeout, dependency, convention, docs, logs, or tool catalog decisions.COMMANDS and NEXT_ACTION over reading or repeating long workflow prose.task_start.py or bootstrap_agent_run.py has emitted run.repo_tool_routing_policy, execute the selected skill tool route in manifest order: show_skill_packet, required_commands, task-matching conditional commands, then validation commands. If a related skill becomes active later, regenerate that skill packet before handoff.Use when agent-side working philosophy, interaction lessons, task retrospectives, repeated routing misses, missed skill invocation, or recurrence-prevention feedback should be logged without mixing them into user preferences.
Use when exploring, refactoring, or choosing an algorithm under proof obligations; builds JIT-canonical IR, lemma dependency graphs, algorithmic blocker frontiers, and algorithm-change guidance before handing terminal proof work to formal-proof-workflow.
Use when designing, implementing, reviewing, or diagnosing numerical optimization, solvers, preconditioners, convergence, gradients, Jacobians, Hessians, KKT conditions, tolerances, or optimization benchmarks; fixes the mathematical and validation contract before code or experiment changes.
Use when a change needs oracle/spec-risk classification or resilient, adversarial static test design, including behavior contracts, oracle choice, property/metamorphic candidates, mutation adequacy, or brittle-test diagnosis.
Use when updating AgentCanon itself, refreshing a vendored vendor/agent-canon submodule pin, repairing AgentCanon root runtime views, applying AgentCanon update TODOs, or routing local AgentCanon source commits through a proper AgentCanon branch and PR before parent pin updates.
Use when accumulated AgentCanon eval evidence is missing, stale, or failing; runs registered eval producers, validates eval family accumulation, and stores evidence through the log archive instead of hand-writing reports.