CCF Common

Family File Contract

Before writing, resolve the canonical output and one stable working directory per task/artifact. Reuse explicit or established task paths; otherwise use project-root ccfa-workfiles/<purpose>/<artifact-id>/, with source/, assets/, cache/, and build/ only as needed. Update current files in place; do not scatter intermediates or create iteration copies. Preserve inputs and required evidence; clean only verified disposable files created by this task. Use UTF-8 text I/O and check Chinese text after saving or rendering. For file work, apply artifact-contracts.md and reuse the same paths across skill transitions.

Collaboration Contract

Before applying these shared controls, activate ccf-humanization unless its baseline is already active. Apply this entry once, then continue specialist work; do not recursively re-enter either preflight. Reuse applicable rules across contributors.

Keep one integrating owner and actively use other skills to resolve missing prerequisites or check material findings. Reuse applicable evidence; do not skip necessary groundwork to save tokens. Before finalizing, integrate contributions and verify affected results. Follow the conditional cooperation routes; avoid unrelated stages and duplicate reports.

Core Rule

This is the shared control module activated before every CCFA specialist, after ccf-humanization. It sets the task's scope, integrating owner, relevant prerequisites, evidence boundaries, and file paths while specialist skills produce research deliverables. Applying these controls does not run the maintenance workflow or require a governance report.

Invocation Controls

CCFA Handoff Mode: PARTIAL (Recommended). Follow metadata.ccf_skill_controls.handoff_question_mode and references/handoff-modes.md. On a direct invocation, ensure Humanization's baseline has been read before applying these controls. Do not recursively restart either preflight. Reuse applicable rules across contributors and read only the shared references needed for the current decision.

Family Preflight

Resolve the requested result, scope/permissions, integrating owner, and missing or reusable prerequisites. Use references/routing.md for collaboration and references/handoff-modes.md for transitions. Apply evidence/privacy rules when handling sources or private material, artifact rules before file work, and review standards when assessing. Reuse loaded rules and known paths; create no intake form or state file merely to activate this skill. Continue the specialist task once the applicable controls are established. Reserve the maintenance workflow below for actual family-maintenance requests.

Discovery metadata and a named preflight are not evidence that its instructions were read. Apply the visible entry content, and restore missing or truncated relevant sections through targeted reads. When diagnosing a reported trigger failure, distinguish the installed version, actual tool reads, selected mode, and resulting behavior; do not invent a loading receipt. Ordinary tasks need no startup report or persistent trace.

Resolve preferences once with this skill's scripts/preferences.py resolve, adding --project <established-root> when a project is known; reuse the compact result across handoffs. No file means defaults, with no write or question. Current explicit instructions override project, user, and built-in defaults. These choices govern optional extra work only; necessary checks and preflights remain active. Read only relevant sections of preferences.md for enhancement triggers, first-time choices, configuration changes, or a missing resolver/error; do not load its whole guide or launch the web page on every task.

Keep the current user goal, justified route, and next action distinct. Explore different mechanisms when the task calls for discovery; preserve the selected route during execution unless a material reason changes it. Distinguish proposed, executed, and verified work. After lost context or repeated drift, reconcile actual artifacts and take one bounded step without lowering the deliverable standard. Repeated work needs a new observation, changed hypothesis, or concrete repair. For long tasks, stalled progress, or material recovery, load continuity.md; short tasks need no extra file or review call.

An explicit request for autonomous research or repeated experiments routes to the orchestrator's research loop. It uses the task's experiment scope and budget, not the optional maintenance-candidate allowance. Ordinary writing, concept review, and planning do not start experiments.

Shared Controls

Load only the references needed for the current preflight decision or maintenance task:

• references/routing.md: Use to resolve which CCFA skill owns a request and to avoid trigger overlap.
• references/task-modes.md: Use for exploratory/quick/standard depth, authorization-aware continuation, selective context, host-supported parallel work, and GPT-6 adaptation.
• references/review-output-standards.md: Use to keep numeric scoring, multi-reviewer panels, score-change conditions, and visible output quality consistent.
• references/handoff-modes.md: Use to interpret metadata.ccf_skill_controls.handoff_question_mode.
• references/preferences.md: Resolve remembered optional enhancements, precedence, and the local three-language configuration page; mandatory preflights and necessary checks remain active.
• references/continuity.md: Preserve task intent through rerouting and recovery; use compact state and optional record validation only when useful.
• references/privacy-and-evidence.md: Use when handling manuscripts, reviews, rebuttals, private drafts, literature searches, or evidence claims.
• references/source-registry.yaml: Use as the shared source inventory for venue rules, review methods, exemplar records, and research-workflow references.
• references/ccf-a-venue-map.md: Use when a non-writing skill needs venue-family mapping without depending on ccf-paper-writer.
• references/skill-trigger-registry.yaml: Use as the canonical trigger registry; compare descriptions and agent prompts with the same ownership boundaries.
• references/artifact-contracts.md: Use whenever a skill creates or revises files; it defines ownership, canonical paths, minimal intermediate-file generation, overwrite-in-place defaults, task-scoped cleanup, and required retained history.
• references/ccfa-yaml-contract.md: Use for the shared ccfa.yaml project-state schema.

Maintenance Workflow

1. Follow the current user scope and existing authorization before defaults; for no-new-files maintenance, edit existing surfaces only. When editing any CCFA family skill, preserve the metadata.ccf_skill_controls block and keep its keys aligned with references/handoff-modes.md.
2. Use references/routing.md before adding new trigger language to prevent overlapping ownership.
3. Use references/task-modes.md before changing checklist strictness, quick polishing, standard review, or output contracts.
4. Use references/review-output-standards.md before changing scorecards, reviewer panels, score-risk language, or final output self-check rules.
5. Use references/privacy-and-evidence.md before adding any browsing, citation, novelty, scoring, experiment-result, compression, or rebuttal instruction.
6. Use references/artifact-contracts.md before changing generated filenames, revision behavior, report folders, caches, attempt archives, or ledgers. Default to one canonical artifact per deliverable, one stable working directory per task/artifact, and cleanup of task-created disposable files before delivery. Preserve required evidence and other tasks' files.
7. Never commit personal absolute paths, usernames, expanded home directories, private local skill roots, or machine-specific command examples. Use $CODEX_HOME, $HOME, repo-relative paths, or non-identifying placeholders.
8. Put new public sources in references/source-registry.yaml; do not duplicate long URL lists in sibling source-notes.md files. Local references must use repo-relative or non-identifying local:/repo: identifiers, not machine paths.
9. Run scripts/check_sources.py after source-registry edits. The script reports issues only and must not rewrite registry files.
10. Run scripts/check_v04.py for family metadata, registry, script syntax, referenced resources, and regression checks; then run scripts/check_path_privacy.py before finalizing CCFA-family changes that touch docs, examples, source records, scripts, diagrams, or release files.

Output Contract

For ordinary preflight, return control to the specialist without a separate report. When auditing or updating CCFA skills, report:

Routing impact:
Handoff mode impact:
Private-material safety:
Source-registry changes:
Score-risk language changes:
Validation result: