3.9 KiB

Raw Blame History

Agent Invocation Packet: giant-cognition

1. Invocation Identity

project_id: 水晶裹尸布与生命体
workflow: article-outline-review
stage: outline-review
run_id: r1.2
participant_id: giant-cognition
target_participant_name: 巨人
participant_type: ccpe_lite
canonical_artifact_path: C:\Users\wangq\Documents\Codex\ccpe-system\agents\lite\giant-cognition.prompt.md
canonical_artifact_version: current
invocation_mode: prompt_path_reference
packet_type: participant_startup_packet

2. Carrier Metadata

carrier_type: agent_subsession_participant
thread_id: not_applicable_platform_subagent
agent_id: 019ee35c-8683-7760-9475-3ea3bb3454b2
persistent_required: true
resume_or_followup_method: multi_agent_v1.send_input with recorded agent_id
execution_status: returned
strict_invocation_required: true
simulation_only: false
returned_output_path: C:\Users\wangq\Documents\Codex\writing-workbench\projects\水晶裹尸布与生命体\audit\巨人\r1.2\returned-output.md
skill_execution_record_path: not_applicable
prompt_to_send_path: C:\Users\wangq\Documents\Codex\writing-workbench\projects\水晶裹尸布与生命体\audit\巨人\r1.2\prompt-to-send.md

3. Canonical Artifact Handling

should_paste_full_prompt_or_spec: no
role_integrity_requirement: must_load_canonical_artifact_before_task
allowed_wrapper_instructions: project_id, workflow, stage, run_id, input_files, context_files, output_path, output_language_policy
forbidden_role_changes: replacing the canonical role; replacing the canonical tone; replacing the canonical output structure; treating the task prompt as the role definition
no_simulation_requirement: true
path_readability_evidence: canonical artifact and prior run were readable in r1.1; continuation packet prepared on 2026-06-20T13:43:36+08:00

4. Current Task

继续上一轮巨人评审子会话，评审《水晶裹尸布与生命体提纲4.0.md》。提纲 4.0 是本轮新增评审对象。背景资料已在 r1.1 会话中提供，本轮默认沿用既有上下文，不要求重新通读；如需核对事实边界，可按需重读。上一轮巨人 r1.1 返回用于延续自己的判断，不要求复述。

5. Input Files

C:\Users\wangq\Documents\Codex\writing-workbench\projects\水晶裹尸布与生命体\outline\水晶裹尸布与生命体提纲4.0.md

6. Context Files

C:\Users\wangq\Documents\Codex\writing-workbench\projects\水晶裹尸布与生命体\audit\巨人\r1.1\returned-output.md

Available on demand for fact-boundary checks:

C:\Users\wangq\Documents\Codex\writing-workbench\projects\水晶裹尸布与生命体\intake\背景资料.md

7. Embedded Inputs

No embedded content. Use the local file paths above.

8. Output Contract

source_primary_language: Chinese
output_language_policy: mirror_source_primary_language
human_review_language_requirement: Chinese
human_confirmation_artifact: false
gate_id_if_applicable: not_applicable
expected_output_type: outline_v4_review

9. Continuity Policy

This run continues 巨人's first reviewer session. Preserve the same agent_id for future follow-up rounds such as r1.3 if the user asks 巨人 to continue.

10. Session Logging

The controller records this packet, prompt-to-send.md, run-metadata.json, agent_id, and returned-output.md under this run directory.

11. Forbidden Actions

Do not simulate another reviewer.
Do not write a main-session synthesis.
Do not generate article正文.
Do not modify the outline file.
Do not overwrite files outside this run directory.

12. Return Path

Write the formal returned output to:

C:\Users\wangq\Documents\Codex\writing-workbench\projects\水晶裹尸布与生命体\audit\巨人\r1.2\returned-output.md

13. Runtime Acceptance Rule

The output is formal only if the existing real child agent continued the participant role, returned a result, and returned-output.md records simulation_only: false.

3.9 KiB Raw Blame History Unescape Escape