paperclip-control-plane
Deterministic read-only Paperclip issue, document, and approval-link verification with strict JSON output.
Skill by alvarosanchez
Chalk Labs
Need help deploying paperclip-control-plane?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/alvarosanchez-paperclip-control-plane
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: documents
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
- Runs as
- Node.js package
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/paperclip-control-plane
What's inside
- Skill name
paperclip-control-plane- Description
- Deterministic read-only Paperclip issue, document, and approval-link verification with strict JSON output.
- License
- MIT
- Instructions
- 51 lines in SKILL.md · view on GitHub
Before installing
- Ships executable scripts
- Instructions call the network
Files (4)
- SKILL.md
- instruction-automation-audit.md · references/
- paperclip-workflow.mjs · scripts/
- plugin-tool.mjs · scripts/
SKILL.md (first lines)
# Paperclip Control Plane
Use the bundled CLI before interpreting repeated REST steps:
```bash
node <skill-directory>/scripts/paperclip-workflow.mjs snapshot --issue "$PAPERCLIP_ISSUE_ID" --document qa-intake
node <skill-directory>/scripts/paperclip-workflow.mjs verify --issue "$PAPERCLIP_ISSUE_ID" --participant "$PAPERCLIP_AGENT_ID"
```
Plugin tools go through the same skill so no run has to rediscover the gateway payload:
```bash
node <skill-directory>/scripts/plugin-tool.mjs list
node <skill-directory>/scripts/plugin-tool.mjs call paperclip-github-plugin:get_issue --params '{"repository":"micronaut-projects/micronaut-maven-plugin","issueNumber":1712}'
```
`plugin-tool.mjs` builds the exact `POST /api/plugins/tools/execute` body the host validates, `{ "tool": "<pluginId>:<tool>", "parameters": { ... }, "runContext": { "agentId", "runId", "companyId", "projectId" } }`, from `PAPERCLIP_AGENT_ID`, `PAPERCLIP_RUN_ID`, `PAPERCLIP_COMPANY_ID`, and `PAPERCLIP_PROJECT_ID` (falling back to the current issue's project when that variable is empty). The field is `parameters`, never `arguments`. Exit code `2` means the gateway denied the call (`deny_default`: a missing tool-access profile, report it as a deployment gap) or the tool itself returned an error; do not write an ad hoc curl or shell helper for the gateway.
Read every stage artifact in one call instead of guessing document routes:
```bash
node <skill-directory>/scripts/paperclip-workflow.mjs docs --issue "$PAPERCLIP_TASK_ID" --dir "$PAPERCLIP_RUN_SCRATCH_DIR/docs"
```
`docs` writes each issue document to `<dir>/<key>.md` (all documents, or only the `--document` keys given), prints the paths, byte sizes and latest revision ids as JSON, and exits `2` when a requested key is missing. Documents are addressed by `key` (`/api/issues/{id}/documents/{key}`), never by document id, and the snapshot's `documents` field is an array of `{id, key, title, ...}` objects, not a map. Read the written files in full with your file tool; do not `head` a 15 KB artifact and call it read.
Commands:
- `snapshot`: fetch issue, heartbeat context, and selected durable documents as normalized JSON.
- `verify`: assert expected status, assignee, active participant, stage outcome, and document presence.
- `approval-link`: verify that an approval is linked to the current issue.
- `docs`: write the issue's documents to files for reading.
Run environment you already have (no `env` dump is needed, and never paste the key into a command): `PAPERCLIP_API_URL`, `PAPERCLIP_API_KEY` (reference it as `$PAPERCLIP_API_KEY`), `PAPERCLIP_AGENT_ID`, `PAPERCLIP_RUN_ID`, `PAPERCLIP_COMPANY_ID`, `PAPERCLIP_PROJECT_ID`, `PAPERCLIP_TASK_ID` (the current issue id), `PAPERCLIP_RUN_SCRATCH_DIR` (per-run scratch, deleted after the run; draft artifacts here, never inside the repository worktree), and `PAPERCLIP_WORKSPACE_CWD` (the worktree). File tools cannot expand variables: `echo "$PAPERCLIP_RUN_SCRATCH_DIR"` once and use the literal path.
Write endpoints this package's roles use (authority still comes from the role and stage, not from this list; send `X-Paperclip-Run-Id: $PAPERCLIP_RUN_ID` on every write so the change is attributed to your run, and always print the HTTP status and body):
- durable artifact: `PUT /api/issues/{id}/documents/{key}` with `{"title": "...", "format": "markdown", "body": "...", "changeSummary": "..."}`; an existing key additionally requires `"baseRevisionId"` (the latest revision id from `docs`/`snapshot`). On this raw endpoint a stale `baseRevisionId` or a locked key returns `409`; a native document tool may instead redirect the write to a new key (see below). Either way stop and record the outcome; do not retry or accept a remapped key.
- stage decision: `PATCH /api/issues/{id}` (with the run-id header) and `{"status": "done", "comment": "<decision>"}` approves your active execution-policy stage in one call; a non-`done` status (`in_progress`) with the same `comment` field requests changes. The path parameter is `{id}` (`/api/issues/{id}`), not `{issueId}`.
- plain progress comment: `POST /api/issues/{id}/comments` with `{"body": "..."}`.
- plugin tools: `plugin-tool.mjs` above.
All calls require `PAPERCLIP_API_URL` and `PAPERCLIP_API_KEY`. The URL must be an origin with no credentials, path, query, or fragment; plaintext HTTP is accepted only on loopback. Options are command-specific, singleton options reject duplicates, and unknown options fail before any request. Treat exit code `2` as a failed state assertion, not permission to improvise a write.
Do not attempt a cross-agent heartbeat invocation. Agent-authenticated callers may invoke only themselves; correct execution-policy routing or assignment must wake the next participant. If routing is correct but no run is queued, record a runtime wake blocker.
This skill grants no document mutation, GitHub, repository, implementation, closure, or publication authority; `plugin-tool.mjs` only forwards to tools the gateway already allows for the calling agent. Permitted document writes and stage decisions remain on native Paperclip tools because v2026.831.1 still exposes no server-enforced precondition for same-agent stage re-entry, and an agent write to a locked keyed document is redirected to a new document instead of failing. If a native document operation cannot guarantee the requested key, stop instead of retrying a remapped write.
See `references/instruction-automation-audit.md` for automated and deferred candidates.From alvarosanchez/micronaut-agent-company, MIT.
More from alvarosanchez/micronaut-agent-company
- 0.549score
micronaut-repo-operations
alvarosanchez
Compact router for Micronaut repository workflow, delivery gates, deterministic evidence, and on-demand operating references.
SkillBFreshness: B★ 1Claude Code +9 - 0.549score
ceo-issue-history
alvarosanchez
Build deterministic, complete 30-day issue-level evidence and rank recurring operating-system improvements for the CEO routine.
SkillBFreshness: B★ 1Claude Code +9 - 0.544score
marketplace-skill-discovery
alvarosanchez
Search-only Skills marketplace discovery for CEO Training; inspect candidates and prepare evidence-backed approval requests without installing, updating…
SkillBFreshness: B★ 1Claude Code +9 - 0.544score
product-discovery
alvarosanchez
Run Product Manager discovery as either a coordinator routine that creates self-contained project subtasks or a project-specific subtask that performs deep…
SkillBFreshness: B★ 1Claude Code +9 - 0.544score
micronaut-graalvm-native-development
alvarosanchez
Diagnose, plan, implement, verify, and review Micronaut-specific GraalVM native-image work across Native Build Tools, reachability metadata, micronaut-graal…
SkillBFreshness: B★ 1Claude Code +9 - 0.544score
micronaut-security-review
alvarosanchez
Security review checklist for Micronaut source code, dependencies, build logic, CI/CD, release automation, and secure-default changes.
SkillBFreshness: B★ 1Claude Code +9 - 0.544score
micronaut-test-resources-provider-development
alvarosanchez
Develop, plan, verify, document, review, or secure Micronaut Test Resources providers, including resolver lifecycle, service-loader registration, build-tools…
SkillBFreshness: B★ 1Claude Code +9 - 0.541score
micronaut-quality-gates
alvarosanchez
Shared definition of done for Micronaut planning, implementation, QA, security review, code review, and PR handoff.
SkillBFreshness: B★ 1Claude Code +9 - 0.541score
micronaut-github-operations
alvarosanchez
Shared GitHub, GitHub Sync plugin, footer, PR-linking, and review-thread protocol used by Micronaut company agents.
SkillBFreshness: B★ 1Claude Code +9 - 0.540score
company-package-evolution
alvarosanchez
Decide when CEO self-improvement should stay in additive runtime guidance versus becoming a PR against the Micronaut Agent Company source repository.
SkillBFreshness: B★ 1Claude Code +9 - 0.517score
gh-cli
alvarosanchez
Reference the upstream GitHub gh CLI skill only for explicit human/operator exceptions where a non-plugin GitHub client is authorized.
SkillBFreshness: B★ 1Claude Code +9 - 0.514score
skill-creator
alvarosanchez
Reference the upstream Micronaut project-template skill for creating and refining portable agent skills used by this company package.
SkillBFreshness: B★ 1Claude Code +9
Signals
- Stars
- 1
- Forks
- 0
- Open issues
- 1
- Last commit (this folder)
- Sep 15, 2026 (1 month ago)
- Last release
- Sep 22, 2026 (1 month ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from alvarosanchez
- 0.549score
ceo-issue-history
alvarosanchez
Build deterministic, complete 30-day issue-level evidence and rank recurring operating-system improvements for the CEO routine.
SkillBFreshness: B★ 1Claude Code +9 - 0.549score
micronaut-repo-operations
alvarosanchez
Compact router for Micronaut repository workflow, delivery gates, deterministic evidence, and on-demand operating references.
SkillBFreshness: B★ 1Claude Code +9 - 0.544score
micronaut-test-resources-provider-development
alvarosanchez
Develop, plan, verify, document, review, or secure Micronaut Test Resources providers, including resolver lifecycle, service-loader registration, build-tools…
SkillBFreshness: B★ 1Claude Code +9 - 0.544score
marketplace-skill-discovery
alvarosanchez
Search-only Skills marketplace discovery for CEO Training; inspect candidates and prepare evidence-backed approval requests without installing, updating…
SkillBFreshness: B★ 1Claude Code +9 - 0.544score
micronaut-graalvm-native-development
alvarosanchez
Diagnose, plan, implement, verify, and review Micronaut-specific GraalVM native-image work across Native Build Tools, reachability metadata, micronaut-graal…
SkillBFreshness: B★ 1Claude Code +9 - 0.544score
micronaut-security-review
alvarosanchez
Security review checklist for Micronaut source code, dependencies, build logic, CI/CD, release automation, and secure-default changes.
SkillBFreshness: B★ 1Claude Code +9
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get paperclip-control-plane running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
paperclip-control-plane skill. Agent Mart. https://agentmart.dev/skills/alvarosanchez-paperclip-control-plane (last updated 2026-10-24)