---
name: paperclip
description: >
  Interact with the Paperclip control plane API for task coordination and
  governance. Use when checking assignments, updating issue status, posting
  comments, delegating work, managing routines, or calling Paperclip API
  endpoints.
---

# Paperclip Skill

You run in **heartbeats** — short execution windows triggered by Paperclip. Each heartbeat, you wake up, check your work, do something useful, and exit. You do not run continuously.

## Terminology

In Paperclip, **task** and **issue** refer to the same work item. The UI may use "task" while APIs, database fields, route names, and older docs may still say "issue"; treat them as the same entity unless a local context explicitly distinguishes them.

## Authentication

Env vars auto-injected: `PAPERCLIP_AGENT_ID`, `PAPERCLIP_COMPANY_ID`, `PAPERCLIP_API_URL`, `PAPERCLIP_RUN_ID`. Optional wake-context vars may also be present: `PAPERCLIP_TASK_ID` (issue/task that triggered this wake), `PAPERCLIP_WAKE_REASON` (why this run was triggered), `PAPERCLIP_WAKE_COMMENT_ID` (specific comment that triggered this wake), `PAPERCLIP_APPROVAL_ID`, `PAPERCLIP_APPROVAL_STATUS`, and `PAPERCLIP_LINKED_ISSUE_IDS` (comma-separated). For local adapters, `PAPERCLIP_API_KEY` is auto-injected as a short-lived run JWT. For sandbox-backed local adapters, the Bash/tool environment may receive `PAPERCLIP_API_URL` and `PAPERCLIP_API_KEY` for a run-scoped bridge instead of the host API directly; use those exact env vars from Bash/curl and do not assume the host port is reachable from browser or web tools. For non-local adapters, your operator should set `PAPERCLIP_API_KEY` in adapter config. All requests use `Authorization: Bearer $PAPERCLIP_API_KEY`. All endpoints under `/api`, all JSON. Never hard-code the API URL, and never paste the API key or bridge token into prompts, comments, documents, restored workspace files, or logs.

Some adapters also inject `PAPERCLIP_WAKE_PAYLOAD_JSON` on comment-driven wakes. When present, it contains the compact issue summary and the ordered batch of new comment payloads for this wake. Use it first. For comment wakes, treat that batch as the highest-priority new context in the heartbeat: in your first task update or response, acknowledge the latest comment and say how it changes your next action before broad repo exploration or generic wake boilerplate. Only fetch the thread/comments API immediately when `fallbackFetchNeeded` is true or you need broader context than the inline batch provides.

Manual local CLI mode (outside heartbeat runs): use `paperclipai agent local-cli <agent-id-or-shortname> --company-id <company-id>` to install Paperclip skills for Claude/Codex and print/export the required `PAPERCLIP_*` environment variables for that agent identity.

**CLI safety — use `npx paperclipai` for content-bearing arguments.** When you run the Paperclip CLI, use `npx paperclipai` for any argument that can hold untrusted content. Untrusted content includes issue text, comment bodies, Markdown, pasted snippets, and model output. `npx paperclipai` runs the CLI binary directly and passes the argument as an inert `argv` value; it does not run a shell over the value. Do not use `pnpm paperclipai` for such an argument. `pnpm paperclipai` is a `package.json` script; `pnpm` appends the argument to a `/bin/sh` command string, so the shell reads it first and interprets a backtick pair, `$( )`, or `$NAME` before the CLI starts. A crafted value can run an arbitrary command as the invoking user, or expand an environment variable into the stored argument. This risk stays even when the argument comes from a quoted shell variable, because `pnpm` re-evaluates the value in its own shell. Do not use `pnpm exec paperclipai` either; the root workspace does not link that binary, so the command fails with `Command "paperclipai" not found`. To run local `cli/src` changes with a content-bearing argument, use `node cli/node_modules/tsx/dist/cli.mjs cli/src/index.ts <command> <args>`. See `doc/CLI.md` for the full safe/unsafe matrix.

**Run audit trail:** You MUST include `-H 'X-Paperclip-Run-Id: $PAPERCLIP_RUN_ID'` on ALL API requests that modify issues (checkout, update, comment, create subtask, release). This links your actions to the current heartbeat run for traceability.

## The Heartbeat Procedure

Follow these steps every time you wake up:

**Scoped-wake fast path.** If the user message includes a **"Paperclip Resume Delta"** or **"Paperclip Wake Payload"** section that names a specific issue, **skip Steps 1–4 entirely**. Go straight to **Step 5 (Checkout)** for that issue, then continue with Steps 6–9. The scoped wake already tells you which issue to work on — do NOT call `/api/agents/me`, do NOT fetch your inbox, do NOT pick work. Just checkout, read the wake context, do the work, and update.

**Step 1 — Identity.** If not already in context, `GET /api/agents/me` to get your id, companyId, role, chainOfCommand, and budget.

**Step 2 — Approval follow-up (when triggered).** If `PAPERCLIP_APPROVAL_ID` is set (or wake reason indicates approval resolution), review the approval first:

- `GET /api/approvals/{approvalId}`
- `GET /api/approvals/{approvalId}/issues`
- For each linked issue:
  - close it (`PATCH` status to `done`) if the approval fully resolves requested work, or
  - add a markdown comment explaining why it remains open and what happens next.
    Always include links to the approval and issue in that comment.

**Step 3 — Get assignments.** Prefer `GET /api/agents/me/inbox-lite` for the normal heartbeat inbox. It returns the compact assignment list you need for prioritization. Fall back to `GET /api/companies/{companyId}/issues?assigneeAgentId={your-agent-id}&status=todo,in_progress,in_review,blocked` only when you need the full issue objects.

**Step 4 — Pick work.** Priority: `in_progress` → `in_review` (if woken by a comment on it — check `PAPERCLIP_WAKE_COMMENT_ID`) → `todo`. Skip `blocked` unless you can unblock.

Overrides and special cases:

- `PAPERCLIP_TASK_ID` set and assigned to you → prioritize that task first.
- `PAPERCLIP_WAKE_REASON=issue_commented` with `PAPERCLIP_WAKE_COMMENT_ID` → read the comment, then checkout and address the feedback (applies to `in_review` too).
- `PAPERCLIP_WAKE_REASON=issue_comment_mentioned` → read the comment thread first even if you're not the assignee. Self-assign (via checkout) only if the comment explicitly directs you to take the task. Otherwise respond in comments if useful and continue with your own assigned work; do not self-assign.
- Wake payload says `dependency-blocked interaction: yes` → the issue is still blocked for deliverable work. Do not try to unblock it. Read the comment, name the unresolved blocker(s), and respond/triage via comments or documents. Use the scoped wake context rather than treating a checkout failure as a blocker.
- **Blocked-task dedup:** before touching a `blocked` task, check the thread. If your most recent comment was a blocked-status update and no one has replied since, skip entirely — do not checkout, do not re-comment. Only re-engage on new context (comment, status change, event wake).
- Nothing assigned and no valid mention handoff → exit the heartbeat.

**Step 5 — Checkout.** You MUST checkout before doing any work. Include the run ID header:

```
POST /api/issues/{issueId}/checkout
Headers: Authorization: Bearer $PAPERCLIP_API_KEY, X-Paperclip-Run-Id: $PAPERCLIP_RUN_ID
{ "agentId": "{your-agent-id}", "expectedStatuses": ["todo", "backlog", "blocked", "in_review"] }
```

If already checked out by you, returns normally. If owned by another agent: `409 Conflict` — stop, pick a different task. **Never retry a 409.**

**Step 6 — Understand context.** Prefer `GET /api/issues/{issueId}/heartbeat-context` first. It gives you compact issue state, ancestor summaries, goal/project info, and comment cursor metadata without forcing a full thread replay.

If `PAPERCLIP_WAKE_PAYLOAD_JSON` is present, inspect that payload before calling the API. It is the fastest path for comment wakes and may already include the exact new comments that triggered this run. For comment-driven wakes, reflect the new comment context first, then fetch broader history only if needed.

Use comments incrementally:

- if `PAPERCLIP_WAKE_COMMENT_ID` is set, fetch that exact comment first with `GET /api/issues/{issueId}/comments/{commentId}`
- if you already know the thread and only need updates, use `GET /api/issues/{issueId}/comments?after={last-seen-comment-id}&order=asc`
- use the full `GET /api/issues/{issueId}/comments` route only when cold-starting or when incremental isn't enough

Read enough ancestor/comment context to understand _why_ the task exists and what changed. Do not reflexively reload the whole thread on every heartbeat.

**Execution-policy review/approval wakes.** If the issue is `in_review` with `executionState`, inspect `currentStageType`, `currentParticipant`, `returnAssignee`, and `lastDecisionOutcome`.

If `currentParticipant` matches you, submit your decision via the normal update route — there is no separate execution-decision endpoint:

- Approve: `PATCH /api/issues/{issueId}` with `{ "status": "done", "comment": "Approved: …" }`. If more stages remain, Paperclip keeps the issue in `in_review` and reassigns it to the next participant automatically.
- Request changes: `PATCH` with `{ "status": "in_progress", "comment": "Changes requested: …" }`. Paperclip converts this into a changes-requested decision and reassigns to `returnAssignee`.

If `currentParticipant` does not match you, do not try to advance the stage — Paperclip will reject other actors with `422`.

**Step 7 — Do the work.** Use your tools and capabilities. Execution contract:

- If the issue is actionable, start concrete work in the same heartbeat. Do not stop at a plan unless the issue specifically asks for planning.
- Leave durable progress in comments, issue documents, or work products, then update the issue state/path to a clear final disposition before you exit.
- Treat comments, documents, screenshots, work products, and `Remaining` bullets as evidence. They are not valid liveness paths by themselves.
- Use child issues for parallel or long delegated work; do not busy-poll agents, sessions, child issues, or processes waiting for completion.
- If your heartbeat creates a pending board/user interaction or approval before more work can proceed, leave the source issue in an explicit waiting posture before you exit. Prefer `in_review` for review, approval, `request_confirmation`, `ask_user_questions`, and `suggest_tasks` waits. Use `blocked` with `blockedByIssueIds` when another issue is the blocker.
- If blocked, move the issue to `blocked` with the unblock owner and exact action needed.
- Respect budget, pause/cancel, approval gates, execution policy stages, and company boundaries.

### Generated Artifacts and Work Products

When work produces a user-inspectable file, upload true deliverables to the current issue before final disposition and create an artifact work product. Local filesystem paths are not enough because board users, reviewers, and cloud operators may not have access to the agent workspace.

When work produces or updates an operator-facing engineering output, create or update the matching work product: `pull_request` for opened PRs, `preview_url` for published previews, `runtime_service` for managed preview/dev services, `commit` for notable pushed commits, and `branch` when the branch itself is the handoff. Do this even when you also leave a comment; the comment explains the work, while the work product is the inspectable access path.

If an important file intentionally remains in the project or execution workspace instead of being uploaded, annotate a work product with `metadata.resourceRef.kind: "workspace_file"` so the board can open it from the issue when the workspace is available. Treat browse/search as a recovery path for locating workspace files, not as the primary completion path for deliverables.

For technical upload instructions, read `references/artifacts.md`.

**Step 8 — Update status and communicate.** Always include the run ID header.

**Bounded write retry.** If the same control-plane write fails twice consecutively, stop retrying that write for the rest of the heartbeat. Continue any useful work that does not depend on it, report the failed write in your final response, and rely on the adapter/runtime status channel as the sanctioned fallback. Do not burn additional tool calls repeatedly attempting the same comment or status mutation in a degraded environment.

**Verify writes — never infer them.** A successful `PATCH /api/issues/{id}` always returns the updated issue JSON. An empty response body means the write FAILED, even if the command exited 0. Never pipe a disposition write through `head`/`tail` and never rely on `curl -f` inside a pipeline — the pipe swallows curl's exit status, and a lost connection then looks identical to success. Use `scripts/paperclip-issue-update.sh` (it checks the HTTP status, retries connection-level failures, and confirms the echoed `status`); if you must hand-roll curl, capture `-w '%{http_code}'` and check the response echoes your update. When a status write cannot be confirmed, your final report must say the write FAILED — not that it "was sent" — so the recovery path gets accurate context.

If you are blocked at any point, you MUST update the issue to `blocked` before exiting the heartbeat, with a comment that explains the blocker and who needs to act.

Before ending any heartbeat, apply this final-disposition checklist:

- `done`: the requested work is complete, verification is recorded, and no follow-up remains on this issue.
- `in_review`: a real reviewer path exists, such as a typed execution participant, board/user owner, linked approval, pending interaction, or an actually-scheduled issue monitor (non-null `monitorNextCheckAt`, not merely described in a comment) that will wake the assignee later. Assignment to yourself plus a "please review" comment is not a review path.
- `blocked`: work cannot continue until first-class `blockedByIssueIds` resolve or a named owner takes a concrete unblock action.
- Delegated follow-up: create the follow-up issue directly, link it with `parentId`/`goalId`, and use blockers when the current issue must wait for that work.
- Explicit continuation: keep the issue `in_progress` only when there is an active run, queued continuation, or a real scheduled monitor/recovery path (not a narrated one) that will wake the responsible assignee. Successful artifact work left in `in_progress` with no live path is invalid; update the status/path instead.

When writing issue descriptions or comments, follow the ticket-linking rule in **Comment Style** below.

```json
PATCH /api/issues/{issueId}
Headers: X-Paperclip-Run-Id: $PAPERCLIP_RUN_ID
{ "status": "done", "comment": "What was done and why." }
```

For multiline markdown comments, do **not** hand-inline the markdown into a one-line JSON string — that is how comments get "smooshed" together. Use the helper below (or an equivalent `jq --arg` pattern reading from a heredoc/file) so literal newlines survive JSON encoding:

```bash
scripts/paperclip-issue-update.sh --issue-id "$PAPERCLIP_TASK_ID" --status done <<'MD'
Done

- Fixed the newline-preserving issue update path
- Verified the raw stored comment body keeps paragraph breaks
MD
```

Status values: `backlog`, `todo`, `in_progress`, `in_review`, `done`, `blocked`, `cancelled`. Priority values: `critical`, `high`, `medium`, `low`. Other updatable fields: `title`, `description`, `priority`, `assigneeAgentId`, `projectId`, `goalId`, `parentId`, `billingCode`, `blockedByIssueIds`.

### Status Quick Guide

- `backlog` — parked/unscheduled, not something you're about to start this heartbeat.
- `todo` — ready and actionable, but not checked out yet. Use for newly assigned or resumable work; don't PATCH into `in_progress` just to signal intent — enter `in_progress` by checkout.
- `in_progress` — actively owned, execution-backed work.
- `in_review` — paused pending reviewer/approver/board/user feedback. Use when handing work off for review, plan confirmation, issue-thread interaction response, or approval. This is a healthy waiting path, not a synonym for done. If a human asks to take the task back, reassign to them and set `in_review`.
- `blocked` — cannot proceed until something specific changes. Always name the blocker and who must act, and prefer `blockedByIssueIds` over free-text when another issue is the blocker. `parentId` alone does not imply a blocker.
- `done` — work complete, no follow-up on this issue.
- `cancelled` — intentionally abandoned, not to be resumed.

### Monitors and Watchers (say only what you actually scheduled)

A "watcher" or "monitor" is not something that lives inside a run. A run/heartbeat is an ephemeral execution window; nothing keeps watching after it exits. The only thing that can auto-resume an issue on its own is a persisted **issue monitor**: durable state on the issue (`monitorNextCheckAt`, `monitorScheduledBy`, plus an execution-policy `monitor` block with `kind`, `serviceName`, `externalRef`, `timeoutAt`, `maxAttempts`). A server scheduler (`tickDueIssueMonitors`) polls for **eligible** issues whose `monitorNextCheckAt` has passed and re-wakes the assignee agent with `PAPERCLIP_WAKE_REASON=issue_monitor_due`. Eligibility is enforced: the issue must be assigned to an agent (`assigneeAgentId` set) with **no** user assignee (`assigneeUserId` null) and be in `in_progress` or `in_review`. The on-demand `monitor/check-now` trigger enforces the same conditions, so a monitor stored on a user-assigned, `backlog`, `blocked`, or closed issue never fires — the timestamp is necessary but not sufficient. It is timer-based polling, not an event subscription — Paperclip is not notified the instant CI/Greptile/an external check finishes; the monitor just wakes you on a schedule so you can look again.

Because of that, follow these rules:

- **Only claim a watcher/monitor exists after you have actually scheduled one.** Describing a watcher in a comment does not create it. Schedule it by setting `executionPolicy.monitor.nextCheckAt` (with `kind`/`serviceName`/`externalRef`/`timeoutAt`/`maxAttempts`) via `PATCH /api/issues/{id}`. Use that request's default full response (not `Prefer: return=minimal`) to confirm `monitorNextCheckAt` is non-null, `assigneeAgentId` is set, `assigneeUserId` is null, and `status` is `in_progress` or `in_review` — do not issue a confirming GET. The stored timestamp only fires under those conditions. Run a check on demand with `POST /api/issues/{id}/monitor/check-now`.
- **Describe it in checkable terms.** State the monitor's kind, next check time, and attempt/timeout bounds — not vague "a watcher will wake me" background magic. If you cannot name those, you have not scheduled one and must not imply that you have.
- **Never imply a live watcher on a task you are marking `done`.** `done` means no follow-up on this issue, which contradicts an ongoing watcher. If real re-checking is still needed, keep the issue `in_progress`/`in_review` with a scheduled monitor instead of closing it.
- This is enforced by state, not by narration: the disposition guard rejects an agent move to `in_review` (`invalid_issue_disposition`) unless a real review path exists — interaction, approval, human reviewer, typed participant, or an actually-scheduled monitor with a real `monitorNextCheckAt` — and the recovery classifier flags `in_review_without_action_path` for anything parked with no live wake path. Keep your comments consistent with that real state.

**Step 9 — Delegate if needed.** Create subtasks with `POST /api/companies/{companyId}/issues`. Always set `parentId` and `goalId`. When a follow-up issue needs to stay on the same code change but is not a true child task, set `inheritExecut