manage-reviews
Skill43 starsChanged 4 months agoMITView on GitHub
Manage reviews: internal and client approvals, sign-offs, revisions and reminders. \"who hasn't approved yet\"
In this repository21 other files it gives its agents.
- AGENTS.mdAGENTS.md
- adapt-copySkill
- assemble-documentSkill
- brand-managerSkill
- build-review-gallerySkill
- c2pa-signSkill
- compose-creativeSkill
- create-previewsSkill
- finalize-monthSkill
- full-pipelineSkill
- generate-videoSkill
- ideate-monthSkill
- index-assetsSkill
- ingest-performanceSkill
- match-assetsSkill
- model-checkSkill
- parse-calendarSkill
- price-checkSkill
- render-carouselsSkill
- research-monthSkill
- setupSkill
The file
Files
/socialforge:manage-reviews — Review & Approval Manager
Manage the multi-tier approval workflow per brand's approval-chain.json. The user starts this skill; it is not model-invocable. --client-send and the reminders of --check message people, so they go through the Execution gate below.
Execution gate
Sending is a side effect other people see. Before any Slack message, email or reminder goes out, show an Execution Summary: who receives it (each recipient or channel), through which connector, which posts, and the exact message text. Send only on an explicit yes typed by the user for exactly that summary; any other reply cancels with nothing sent and no status changed. A changed recipient, post list or message is a new summary and a new yes. Hosts that ignore disable-model-invocation (Codex, for one) still get this gate.
Approval Tiers
- HERO content: Internal review → Client approval → CEO approval (if configured)
- HUB content: Internal review → Client approval (optional)
- HYGIENE content: Internal review only (auto-approvable if configured)
Actions
/socialforge:manage-reviews --approve P04— Approve post P04 at current tier/socialforge:manage-reviews --revise P04 "Make the background warmer"— Request revision with feedback/socialforge:manage-reviews --client-send— Send all internally-approved posts to client review/socialforge:manage-reviews --check— Check pending approvals and send reminders
State Transitions
QUEUED → ASSET_MATCHING → GENERATING → PENDING_REVIEW PENDING_REVIEW → APPROVED_INTERNAL | REVISION_REQUESTED | REJECTED REVISION_REQUESTED → GENERATING (re-enters production with feedback) REJECTED → QUEUED (start over) APPROVED_INTERNAL → PENDING_CLIENT (if client review required) | FINAL (if not) PENDING_CLIENT → APPROVED_CLIENT | REVISION_REQ_CLIENT | REJECTED_CLIENT APPROVED_CLIENT → PENDING_CEO (if CEO required) | FINAL (if not) PENDING_CEO → APPROVED_CEO → FINAL FINAL → (write-protected, no transitions)
Rules
- FINAL status is write-protected — cannot be modified after finalization
- Revision requests re-enter the generation pipeline with specific feedback
- Max 3 revision cycles per HERO post before escalation
- Reminders sent via Slack/email after N days (per escalation_rules)
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.